@namzu/sdk 42.0.2 → 44.0.0

package/src/runtime/query/index.ts

@@ -1,4 +1,3 @@
-import { join } from 'node:path'
 import {
 	AdvisorRegistry,
 	AdvisoryContext,
@@ -14,21 +13,23 @@ import type { ContextReducer } from '../../compaction/reducer.js'
 import { serializeState as serializeWorkingState } from '../../compaction/serializer.js'
 import { restoreWorkingState, snapshotWorkingState } from '../../compaction/wire.js'
 import { type CompactionConfig, CompactionConfigSchema } from '../../config/runtime.js'
-import { TOOL_OUTPUT_DIR_NAME } from '../../constants/tools/index.js'
-import { EmergencySaveManager } from '../../manager/run/emergency.js'
-import type { RunPersistence } from '../../manager/run/persistence.js'
+import { childSessionLog } from '../../manager/agent/child-session.js'
+import { type TurnRecorder, recordedTurnStart } from '../../manager/session/turn-recorder.js'
 import { PromptContributionRegistry } from '../../prompt/contributions.js'
 import { resolveProviderCapabilities } from '../../provider/capabilities.js'
 import type { ProviderChainMember } from '../../provider/fallback.js'
 import { withStreamIdleTimeout } from '../../provider/idle-timeout.js'
 import type { ProviderRetryConfig } from '../../provider/retry.js'
 import { withTokenBudget } from '../../provider/token-budget.js'
-import type { TokenBudget } from '../../run/token-budget.js'
-import type { PathBuilder } from '../../session/workspace/path-builder.js'
+// The session → turn surface QueryParams and query() are frozen against.
+import type { SessionPaths } from '../../session/paths.js'
+import type { SessionTokenBudget, SessionTokenBudgetStore } from '../../store/budget/index.js'
+import type { SessionCheckpointStore } from '../../store/checkpoint/index.js'
+import type { SessionLease, SessionLog } from '../../store/session-log/index.js'
 import {
 	GENAI,
 	NAMZU,
-	agentRunSpanName,
+	agentTurnSpanName,
 	parentContext,
 	serializeSpan,
 } from '../../telemetry/attributes.js'
@@ -53,7 +54,7 @@ import {
 	type ResumeHandler,
 	autoApproveHandler,
 } from '../../types/hitl/index.js'
-import type { CheckpointId, RunId, SessionId, TenantId } from '../../types/ids/index.js'
+import type { CheckpointId, MessageId, SessionId, TenantId } from '../../types/ids/index.js'
 import type { InvocationState } from '../../types/invocation/index.js'
 import type { MemoryStore } from '../../types/memory/index.js'
 import {
@@ -64,24 +65,28 @@ import {
 import type { AgentPersona } from '../../types/persona/index.js'
 import type { LLMProvider } from '../../types/provider/index.js'
 import type { TaskRouterConfig } from '../../types/router/index.js'
-import type { ReviewAnswer } from '../../types/run/answer-review.js'
-import type { CheckpointStore, FencingToken } from '../../types/run/checkpoint-store.js'
-import type { RunEventCursor, RunEventReplay } from '../../types/run/event-cursor.js'
+import type { Sandbox, SandboxProvider } from '../../types/sandbox/index.js'
+import type { ReviewAnswer } from '../../types/session/answer-review.js'
+import type { ProjectId, TopicId } from '../../types/session/ids.js'
 import type {
-	AgentRunConfig,
 	BeforeStep,
 	PrepareStepChain,
-	Run,
-	RunEvent,
-	RunEventListener,
 	StepResult,
 	StopCondition,
-} from '../../types/run/index.js'
-import type { PromoteMemory } from '../../types/run/memory-promotion.js'
-import type { RunStore } from '../../types/run/store.js'
-import type { TokenBudgetStore } from '../../types/run/token-budget-store.js'
-import type { Sandbox, SandboxProvider } from '../../types/sandbox/index.js'
-import type { ProjectId, TopicId } from '../../types/session/ids.js'
+} from '../../types/session/index.js'
+import {
+	type Origin,
+	type SessionEvent,
+	type SessionEventListener,
+	type SessionLogCursor,
+	type SessionLogReplay,
+	type Turn,
+	type TurnConfig,
+	type TurnForkOrigin,
+	type TurnId,
+	isTurnInProgressError,
+} from '../../types/session/index.js'
+import type { PromoteMemory } from '../../types/session/memory-promotion.js'
 import type { Skill } from '../../types/skills/index.js'
 import type { StructuredOutputConfig } from '../../types/structured-output/index.js'
 import type { TaskStore } from '../../types/task/index.js'
@@ -96,7 +101,8 @@ import { AwaitedJobs } from '../jobs/awaited-jobs.js'
 import type { BackgroundJobRegistry } from '../jobs/registry.js'
 import { catchUpFromCursor, settlePreStartCancellation } from './cancelled-before-start.js'
 import { CheckpointManager, findPendingCheckpoint } from './checkpoint.js'
-import { finalizeRun } from './finalize-run.js'
+import type { EventTranslator } from './events.js'
+import { finalizeTurn } from './finalize-turn.js'
 import { GuardCoordinator } from './guard.js'
 import { runInputGuardrails } from './guardrails.js'
 import { IterationOrchestrator } from './iteration/index.js'
@@ -105,17 +111,17 @@ import { isWorkingMemoryMessage } from './iteration/phases/working-memory.js'
 import { applyLifecycleHookResults } from './plugin-hooks.js'
 import {
 	type SelectedResumeState,
-	prepareRun,
+	prepareTurn,
 	projectStateBearingHistory,
 	resolveProviderContextWindow,
 	selectedResumeStates,
-} from './prepare-run.js'
+} from './prepare-turn.js'
 import type { ProjectInstructionContext } from './project-instructions.js'
 import type { PromptCache } from './prompt-cache.js'
 import { PromptBuilder } from './prompt.js'
 import type { PromptSegments } from './prompt.js'
 import { PendingAnswers, QuestionParkBinding } from './question-park.js'
-import { releaseRunResources } from './release-run.js'
+import { releaseTurnResources } from './release-turn.js'
 import { RepeatCallTracker } from './repeat-call.js'
 import { ResultAssembler } from './result.js'
 import {
@@ -137,10 +143,14 @@ import { ToolingBootstrap } from './tooling.js'
 export interface QueryParams {
 	/** Share observations across turns of one conversation and filesystem; otherwise run-local. */
 	fileReadTracker?: import('../../types/tool/index.js').FileReadTracker
-	/** One account shared with the task scheduler and descendant runs. */
-	budget?: TokenBudget
-	/** Canonical tree ledger; defaults to disk beside the root run. */
-	tokenBudgetStore?: TokenBudgetStore
+	/** One ledger shared with the task scheduler and descendant child sessions, keyed by (rootSessionId, rootTurnId). */
+	budget?: SessionTokenBudget
+	/**
+	 * Where root-turn ledgers are kept (`<root-session-id>/budgets/<root-turn-id>.json`).
+	 * Absent: beside the session under {@link QueryParams.paths}, or in memory
+	 * for an in-memory {@link QueryParams.sessionLog}.
+	 */
+	tokenBudgetStore?: SessionTokenBudgetStore
 	/**
 	 * Notice when the model issues the identical tool call repeatedly, and
 	 * say so on the next `tool_result`. Defaults on.
@@ -152,7 +162,7 @@ export interface QueryParams {
 	repeatCallAdvisory?: boolean
 
 	/**
-	 * Tool names this run may not use, subtracted from its effective list.
+	 * Tool names this turn may not use, subtracted from its effective list.
 	 * See {@link import('../../types/agent/base.js').BaseAgentConfig.deniedTools}.
 	 */
 	deniedTools?: readonly string[]
@@ -164,7 +174,7 @@ export interface QueryParams {
 	provider: LLMProvider
 	/**
 	 * Transient-failure policy for model calls. A single 429 or 503 used to
-	 * terminate a run outright — no driver in the estate retries. Defaults
+	 * terminate a turn outright — no driver in the estate retries. Defaults
 	 * to {@link DEFAULT_PROVIDER_RETRY}; pass `false` to opt out (e.g. when
 	 * the host already wraps the provider with its own policy).
 	 *
@@ -190,38 +200,18 @@ export interface QueryParams {
 	 */
 	fallbackProviders?: readonly ProviderChainMember[]
 
-	/**
-	 * Install process-level crash handlers that dump this run's state to
-	 * `<runDir>/../emergency/<runId>.json` on SIGINT, SIGTERM or an
-	 * uncaught exception. `replay({ fromCheckpoint: 'emergency' })` reads
-	 * that file.
-	 *
-	 * **Off by default, and it must stay that way.** `attach` registers
-	 * `process.on(...)` handlers that call `process.exit()`. A library
-	 * seizing a host's termination path is an overreach in any embedded
-	 * context (an API server has its own drain sequence), and the manager
-	 * is a singleton whose `attach` detaches whoever held it before — so
-	 * with concurrent runs the last one to start would silently become the
-	 * only one that gets saved.
-	 *
-	 * Turn it on for a process the run owns end-to-end: a CLI, a worker
-	 * that handles one run at a time. The handlers are removed when the
-	 * run settles.
-	 */
-	emergencySave?: boolean
-
 	/**
 	 * Durability for questions raised by a tool that closed over its
-	 * binding before the run existed.
+	 * binding before the turn existed.
 	 *
 	 * The built-in `ask_user_question` is built with the agent's tool
 	 * registry, so only whoever builds the tools can hand it one — that is
-	 * what lets a single tool instance be durable inside a run and inert
+	 * what lets a single tool instance be durable inside a turn and inert
 	 * outside one. Without it, THAT tool's park is only a suspended
 	 * `await`: kill the process while somebody is looking at the card and
 	 * the answer can never be applied.
 	 *
-	 * Not required for `ToolContext.requestPause`. The run builds that
+	 * Not required for `ToolContext.requestPause`. The turn builds that
 	 * seam per call and binds its own recorder when none is passed, so a
 	 * pause raised from a host-authored tool is durable on every surface
 	 * rather than only on the one agent class that supplies this.
@@ -242,9 +232,9 @@ export interface QueryParams {
 	 * The registry a re-entered `ask_user_question` reads its answer from.
 	 *
 	 * Same shape, same reason and same limit as {@link questionParks}: it
-	 * exists for a tool that closed over the instance before the run did,
-	 * and without it a resumed run re-asks that tool's question. A pause
-	 * from `ToolContext.requestPause` needs none, because the run fills
+	 * exists for a tool that closed over the instance before the turn did,
+	 * and without it a resumed turn re-asks that tool's question. A pause
+	 * from `ToolContext.requestPause` needs none, because the turn fills
 	 * its own on the resume path.
 	 */
 	pendingAnswers?: PendingAnswers
@@ -252,29 +242,29 @@ export interface QueryParams {
 	/** Default per-tool execution deadline. See {@link ToolDefinition.timeoutMs}. */
 	toolTimeoutMs?: number
 	/**
-	 * Where background jobs this run starts are held, and killed.
+	 * Where background jobs this turn starts are held, and killed.
 	 *
-	 * Host-owned so it can outlive one run — a registry built per run could
-	 * never be the thing that kills a run's jobs when the run is already
-	 * gone. This run's jobs are torn down in the `finally` below; another
-	 * run's are untouched. The registry launches host processes, so a run
+	 * Host-owned so it can outlive one turn — a registry built per turn could
+	 * never be the thing that kills a turn's jobs when the turn is already
+	 * gone. This turn's jobs are torn down in the `finally` below; another
+	 * run's are untouched. The registry launches host processes, so a turn
 	 * that also supplies a {@link sandboxProvider} does not expose it to tools:
 	 * background execution is refused rather than silently bypassing the sandbox.
 	 */
 	backgroundJobs?: BackgroundJobRegistry
 	/**
-	 * Which owner the run's background jobs belong to. Absent, the run id:
-	 * jobs are stopped when the run ends. A host that wants a job to
+	 * Which owner the turn's background jobs belong to. Absent, the turn id:
+	 * jobs are stopped when the turn ends. A host that wants a job to
 	 * outlive the turn that started it — a dev server started in one turn
 	 * and read in the next — passes its session id here and calls
-	 * `backgroundJobs.killOwner(sessionId)` when the session ends; the run
+	 * `backgroundJobs.killOwner(sessionId)` when the session ends; the turn
 	 * then stops nothing at its end and still tells the model when a job
 	 * finishes.
 	 */
 	backgroundJobOwner?: string
 
 	/**
-	 * What else goes in this run's prompt.
+	 * What else goes in this turn's prompt.
 	 *
 	 * `static` and `dynamic` contributions reach the system prompt through
 	 * `PromptBuilder`; `turn` contributions reach the ephemeral trailing
@@ -287,8 +277,8 @@ export interface QueryParams {
 	 * Where the `skill` tool loads from.
 	 *
 	 * Separate from `skills`, which is the LIST that goes in the prompt
-	 * manifest. A run can have the manifest without the tool — that is what
-	 * every run did before the tool existed — and the two are wired
+	 * manifest. A turn can have the manifest without the tool — that is what
+	 * every turn did before the tool existed — and the two are wired
 	 * independently on purpose: a host may want the guidance visible without
 	 * granting a way to pull bodies in mid-run.
 	 */
@@ -297,19 +287,19 @@ export interface QueryParams {
 	/**
 	 * Where a message's stored attachments are resolved from.
 	 *
-	 * Absent is fine for every run whose attachments are inline, which is
-	 * every run that existed before this. A message carrying a ref with no
+	 * Absent is fine for every turn whose attachments are inline, which is
+	 * every turn that existed before this. A message carrying a ref with no
 	 * store REFUSES rather than dropping the attachment.
 	 */
 	attachmentStore?: import('../../store/attachment/index.js').AttachmentStore
 	/**
-	 * Maximum wall-clock time for resolving the run's stored attachments.
+	 * Maximum wall-clock time for resolving the turn's stored attachments.
 	 * Defaults to one minute; `0` retains the prior unbounded wait.
 	 */
 	attachmentResolveTimeoutMs?: number
 
 	/**
-	 * How this run reaches the web.
+	 * How this turn reaches the web.
 	 *
 	 * `fetch` and `search` are independent, and this kernel ships only the
 	 * first — see `connector/web` for why choosing a search backend here
@@ -330,12 +320,12 @@ export interface QueryParams {
 
 	/** Max concurrently-executing concurrency-safe tools in one batch. */
 	maxToolConcurrency?: number
-	/** Per-run cumulative tool attempt limit, including nested calls and retries. Unset is unlimited. Re-supply on resume; durable reservations are never refunded. */
+	/** Per-turn cumulative tool attempt limit, including nested calls and retries. Unset is unlimited. Re-supply on resume; durable reservations are never refunded. */
 	maxToolCalls?: number
 
 	/**
 	 * Model-visible size cap for a single tool result. Over-budget output is
-	 * spilled to the run directory and replaced with a head+tail preview
+	 * spilled to the turn directory and replaced with a head+tail preview
 	 * naming the path, so nothing is lost and tokens are paid only if the
 	 * agent decides the rest is worth re-reading. Set `0` to disable.
 	 */
@@ -344,9 +334,9 @@ export interface QueryParams {
 	 * Screens to run against every tool result, where the registry was not
 	 * built with its own.
 	 *
-	 * This is the run's half of a boundary whose only other door is the
+	 * This is the turn's half of a boundary whose only other door is the
 	 * registry constructor — and a registry is usually the HOST's, assembled
-	 * before the run exists, so a run-config option is the only way a run
+	 * before the turn exists, so a turn-config option is the only way a turn
 	 * screens a registry it did not build. A registry built WITH
 	 * `resultGuardrails` states its own policy and wins, `[]` included.
 	 *
@@ -392,7 +382,7 @@ export interface QueryParams {
 	 *
 	 * Before this the only halt was `GuardCoordinator`, which sees four
 	 * numeric budgets and never the messages — so a terminal
-	 * `submit_answer` tool could not end a run, and the model had to be
+	 * `submit_answer` tool could not end a turn, and the model had to be
 	 * prompt-begged to stop with `maxIterations: 200` as the only backstop.
 	 *
 	 * Helpers: `stepCountIs`, `hasToolCall`, `anyOf`.
@@ -400,30 +390,30 @@ export interface QueryParams {
 	stopWhen?: StopCondition
 
 	/**
-	 * Judge the answer the run is about to settle with, and hand it back
+	 * Judge the answer the turn is about to settle with, and hand it back
 	 * with feedback when it is not good enough.
 	 *
 	 * `stopWhen` is only consulted after tools have run, so there was no
-	 * seam at the point the model stops calling them: the run finalized
+	 * seam at the point the model stops calling them: the turn finalized
 	 * with whatever it had. Verify-then-fix — run the build, feed the
-	 * failure back, let it try again — meant starting a new run and
+	 * failure back, let it try again — meant starting a new turn and
 	 * re-supplying the context the first one had already assembled.
 	 *
 	 * Bounded by {@link maxAnswerReviews}. Never called on the forced-final
 	 * turn, which exists to extract a closing summary under pressure.
-	 * Exceptions or malformed verdicts fail the run; cancellation stops waiting.
+	 * Exceptions or malformed verdicts fail the turn; cancellation stops waiting.
 	 */
 	reviewAnswer?: ReviewAnswer
 
 	/**
-	 * Decide what this run should leave behind when it settles.
+	 * Decide what this turn should leave behind when it settles.
 	 *
 	 * See {@link PromoteMemory}. Absent means nothing is offered and the
 	 * run behaves exactly as it did.
 	 */
 	promoteMemory?: PromoteMemory
 
-	/** Corrections allowed before the run stops. Nonnegative safe integer; default 3. Consumed rejections survive checkpoints. */
+	/** Corrections allowed before the turn stops. Nonnegative safe integer; default 3. Consumed rejections survive checkpoints. */
 	maxAnswerReviews?: number
 
 	/** Called with each completed step, as it completes. */
@@ -433,11 +423,11 @@ export interface QueryParams {
 	 * Shape each step before the model is called: narrow the tool surface,
 	 * swap the model, add one-step guidance, change sampling.
 	 *
-	 * `stopWhen` let a run decide TO STOP from what its steps produced;
+	 * `stopWhen` let a turn decide TO STOP from what its steps produced;
 	 * this is the other half — deciding how the next step should look.
 	 * Without it, the tool surface and model are fixed at `query()` time,
 	 * so a phased agent (research with search tools, write with file tools,
-	 * verify with a cheaper model) had to be three separate runs, each
+	 * verify with a cheaper model) had to be three separate turns, each
 	 * starting blind to the last one's context.
 	 *
 	 * Narrowing `activeTools` costs a prompt-cache prefix, since tools
@@ -446,7 +436,7 @@ export interface QueryParams {
 	 * `allowed_tools`, and moving `tool_choice` invalidates cached MESSAGE
 	 * blocks too, which is a strictly worse trade for the same effect.
 	 *
-	 * Fails open — a throw leaves the step with the run's configuration.
+	 * Fails open — a throw leaves the step with the turn's configuration.
 	 */
 	prepareStep?: PrepareStepChain
 	/**
@@ -463,7 +453,7 @@ export interface QueryParams {
 	structuredOutput?: StructuredOutputConfig
 
 	/**
-	 * Checks run BEFORE the first model call. A block settles the run as
+	 * Checks run BEFORE the first model call. A block settles the turn as
 	 * `input_guardrail` having spent nothing.
 	 *
 	 * namzu's three tool gates all point one way — they protect the world
@@ -472,7 +462,7 @@ export interface QueryParams {
 	inputGuardrails?: readonly InputGuardrailSpec[]
 
 	/**
-	 * Checks run against the FINAL result. A block settles the run as
+	 * Checks run against the FINAL result. A block settles the turn as
 	 * `output_guardrail`; a `rewrite` replaces the text (so a PII policy
 	 * can redact rather than discard the whole answer).
 	 *
@@ -482,17 +472,46 @@ export interface QueryParams {
 	 */
 	outputGuardrails?: readonly OutputGuardrailSpec[]
 	tools: ToolRegistryContract
-	runConfig: AgentRunConfig
+	turnConfig: TurnConfig
 	allowedTools?: string[]
 	agentId: string
 	agentName: string
 	workingDirectory?: string
 	/**
 	 * Directories besides the working directory the file tools may reach,
-	 * absolute; a sandboxed run binds each read-write. See
+	 * absolute; a sandboxed turn binds each read-write. See
 	 * `ToolContext.additionalDirectories`.
 	 */
 	additionalDirectories?: readonly string[]
+	/**
+	 * What a file tool's path outside `workingDirectory` and
+	 * `additionalDirectories` becomes, on a turn with no sandbox.
+	 *
+	 * `'refuse'` (the default) keeps the boundary a refusal the tool returns.
+	 * `'review'` makes it a question asked before the call runs: the call is
+	 * marked (`ToolCallSummary.escalation.outsidePaths`), routed to the
+	 * turn's review even when its tool only reads and even where an `allow`
+	 * rule or a remembered grant would have covered it, and — once approved
+	 * — reaches exactly the named paths for that call alone. A `deny` rule
+	 * still refuses it. The review's own mode decides it, so a turn whose
+	 * policy approves everything approves these too, on the audit record.
+	 *
+	 * Only tools that declare `pathArgument` are looked at; a sandboxed turn
+	 * is never, because the path is not mounted there to be reached.
+	 */
+	outsideRootAccess?: 'refuse' | 'review'
+	/**
+	 * Whether a call may ask to run outside the turn's sandbox (the shipped
+	 * `bash` tool's `dangerously_disable_sandbox`).
+	 *
+	 * `'refuse'` (the default) refuses every such request. `'review'` routes
+	 * it to the turn's review every time; it runs unconfined only when the
+	 * decision confirms its id (`HITLResumeDecision.confirmedEscalations`),
+	 * which `createReviewHandler` does only after asking a person, or with
+	 * `unattendedSandboxEscape: 'allow'`. Every confirmation and refusal is
+	 * written to the audit trail.
+	 */
+	sandboxEscape?: 'refuse' | 'review'
 	pricing?: ModelPricing
 	enableActivityTracking?: boolean
 	messages: Message[]
@@ -505,7 +524,7 @@ export interface QueryParams {
 	 * out-of-band — typically in a different process.
 	 *
 	 * Recording a park makes the request survive a restart; this is what
-	 * makes the ANSWER survive one. Without it a resumed run repairs the
+	 * makes the ANSWER survive one. Without it a resumed turn repairs the
 	 * unanswered `tool_use` blocks away and lets the model re-decide, so a
 	 * human's "yes, delete that row" degrades into "ask the model again and
 	 * hope it asks for the same thing".
@@ -525,128 +544,142 @@ export interface QueryParams {
 	 * A park is only worth persisting if a human is actually looking at it:
 	 * a programmatic handler answers in microseconds, and the iteration
 	 * gate runs on every iteration, so recording every park unconditionally
-	 * would take a long run from one full-history checkpoint write per
+	 * would take a long turn from one full-history checkpoint write per
 	 * iteration to three. Set `0` to record every park (tests, or a host
 	 * that wants an unconditional audit trail).
 	 */
 	parkRecordDelayMs?: number
 
 	/**
-	 * Span this run should hang off, when it is a delegated one.
+	 * Span this turn should hang off, when it is a delegated one.
 	 *
 	 * A spawned sub-agent is part of its parent's work, and a trace that
 	 * shows the delegation is the whole reason to trace a supervisor at
-	 * all. Absent for a top-level run, which correctly starts its own root.
+	 * all. Absent for a top-level turn, which correctly starts its own root.
 	 */
 	parentSpan?: import('@opentelemetry/api').Span
 
-	/** Session scope for the run. Required — every run is attributed to a Session. */
+	/** Session scope for the turn. Required — every turn is attributed to a Session. */
 	sessionId: SessionId
 
 	/**
-	 * Topic the Session lives under. Required — every run carries the full
-	 * five-layer scope (Tenant → Project → Topic → Session → Run).
+	 * Topic the Session lives under. Required — every turn carries the full
+	 * five-layer scope (Tenant → Project → Topic → Session → Turn).
 	 * Denormalized from `session.topicId`; callers build this alongside
 	 * `sessionId` so the query pipeline never needs a second SessionStore
 	 * round-trip to recover it.
 	 */
 	topicId: TopicId
 
-	/** Long-lived goal scope for the run. Required. */
+	/** Long-lived goal scope for the turn. Required. */
 	projectId: ProjectId
 
 	/** Isolation boundary (Convention #17). Required. */
 	tenantId: TenantId
 
 	/**
-	 * Optional path layout override. Defaults to a {@link DefaultPathBuilder}
-	 * rooted at `{workingDirectory}/.namzu`. First-call filesystem migration
-	 * runs against this builder's root too, so an injected layout never touches
-	 * the fallback working-directory store as a side effect.
+	 * Where the session lives on disk (`<NAMZU_HOME>/projects/<slug>/…`).
+	 * Defaults to `SessionPaths` under `resolveNamzuHome()`. Namzu never writes
+	 * under the working directory.
 	 */
-	pathBuilder?: PathBuilder
+	paths?: SessionPaths
 
 	/**
-	 * Optional checkpoint persistence override. Absent ⇒ iteration
-	 * checkpoints go to the disk layout under the run's output directory
-	 * (today's behavior). A host injects a scope-keyed
-	 * {@link CheckpointStore} (e.g. Postgres-backed) so mid-turn resume
-	 * survives machines that lose their local disk.
+	 * Optional checkpoint persistence override. Absent: checkpoint documents go
+	 * to `<session-id>/checkpoints/`, or stay in memory for an in-memory
+	 * {@link QueryParams.sessionLog}.
 	 */
-	checkpointStore?: CheckpointStore
+	checkpointStore?: SessionCheckpointStore
 
 	/**
-	 * The fence of the claim this worker holds on the run, from `claimRun`.
-	 *
-	 * Presented on every checkpoint the run writes, so a worker that stalled
-	 * past its lease is refused rather than writing into a run somebody else
-	 * has taken over. Omit it for single-writer deployments, which is what
-	 * every run did before claims existed.
-	 *
-	 * This hop did not exist for a release. The claim, the fence and the
-	 * store-side refusal were all built and tested, and no path between a run
-	 * and its store carried the number — so every checkpoint a RUN wrote went
-	 * out unfenced while the tests, which called the store directly, all
-	 * passed. A capability complete except for the wire between its halves
-	 * reads exactly like a working one.
-	 *
-	 * It fences checkpoints and nothing else. {@link QueryParams.runStore}
-	 * takes no fence, so two workers that both took one run still overwrite
-	 * each other's run record, transcript and report — see the changeset.
+	 * The lease this worker holds on the session, from `claimSession`. Every
+	 * record the turn appends carries its fence (`gen`), so a worker that
+	 * stalled past its lease is refused rather than writing into a session
+	 * somebody else has taken over. Absent: `query()` claims the lease itself
+	 * and releases it when the turn settles or parks.
 	 */
-	claimFence?: FencingToken
+	lease?: SessionLease
 
 	/**
-	 * Where this run records its own evidence — the run record, its messages,
-	 * its transcript and its report. Defaults to the disk layout under the
-	 * resolved output directory.
-	 *
-	 * The sibling of {@link QueryParams.checkpointStore}, and it should always
-	 * have been one: checkpoints could be pointed at durable storage and the
-	 * evidence could not.
+	 * The session log this turn appends to. Defaults to the log at
+	 * {@link QueryParams.paths}. An `InMemorySessionLog` with no `paths` keeps
+	 * the whole session in memory: its checkpoints, its ledger and its child
+	 * sessions.
 	 */
-	runStore?: RunStore
+	sessionLog?: SessionLog
 
 	/**
-	 * Where a reconnecting consumer left off, so this run's stream can start by
+	 * Where a reconnecting consumer left off, so this turn's stream can start by
 	 * handing back what it missed.
 	 *
 	 * The case this serves is the one that exists without a network hop: the
-	 * process holding the run died, and the consumer watching it is coming back
-	 * to a run that has to be resumed. Pair it with `resumeFromCheckpoint` — or
-	 * reach it through {@link import('./resume-run.js').resumeRun}, which is the
+	 * process holding the turn died, and the consumer watching it is coming back
+	 * to a turn that has to be resumed. Pair it with `resumeFromCheckpoint` — or
+	 * reach it through {@link import('./resume-session.js').resumeSession}, which is the
 	 * surface that does both — and the missed durable events are yielded, in
-	 * order, before the resumed run emits anything of its own.
+	 * order, before the resumed turn emits anything of its own.
 	 *
-	 * On a run with no log to catch up on the cursor is answered honestly rather
+	 * On a turn with no log to catch up on the cursor is answered honestly rather
 	 * than ignored: a `sinceSeq` above what exists is `cursor_ahead`, not
 	 * silence.
 	 *
 	 * What comes back is message-granular. Streaming deltas are never persisted
-	 * — see {@link import('../../types/run/store.js').RunStore.appendEvent} —
+	 * — see {@link import('../../types/session/events.js').isEphemeralEvent} —
 	 * so a late subscriber recovers the assistant text, the tool results and the
 	 * lifecycle, not the keystroke cadence that produced them.
 	 */
-	eventCursor?: RunEventCursor
+	eventCursor?: SessionLogCursor
 
 	/**
 	 * What became of {@link QueryParams.eventCursor}.
 	 *
 	 * A callback rather than an event on the stream, because the answer is about
-	 * the SUBSCRIPTION and not about the run — and rather than a throw, because
-	 * a stale cursor is a client's problem and must not be able to stop a run
+	 * the SUBSCRIPTION and not about the turn — and rather than a throw, because
+	 * a stale cursor is a client's problem and must not be able to stop a turn
 	 * from continuing. A host that receives `unavailable` re-derives from the
 	 * transcript; one that receives nothing at all would splice a hole into its
 	 * state and never know.
 	 *
-	 * Called once, before the run's first event, and only when a cursor was
+	 * Called once, before the turn's first event, and only when a cursor was
 	 * supplied.
 	 */
-	onEventReplay?: (replay: RunEventReplay) => void
+	onEventReplay?: (replay: SessionLogReplay) => void
 
-	runId?: RunId
+	/**
+	 * Continue this turn (with `resumeFromCheckpoint`). Absent: a new turn is
+	 * begun, and refused with `TurnInProgressError` when the session already
+	 * has an active one.
+	 */
+	turnId?: TurnId
+
+	/**
+	 * Close an `interrupted` active turn — one whose process is gone — with
+	 * `turn_failed{interrupted}` before this turn begins, instead of refusing
+	 * with `TurnInProgressError`. A running or paused turn is never closed
+	 * this way. An interactive host that never resumes an interrupted turn
+	 * sets it.
+	 */
+	abandonInterrupted?: boolean
+
+	/**
+	 * Which protocol opened this turn, and the caller-side ids it used
+	 * (`turn_started.origin`). A protocol adapter records the client's own
+	 * turn id here (`externalTurnId`) rather than making it a namzu id.
+	 */
+	origin?: Origin
+
+	/**
+	 * Present when this session was forked from another session's checkpoint:
+	 * written into `session_started.forkedFrom` when the log is started, and
+	 * carried on the returned turn. See `prepareForkState`.
+	 */
+	forkedFrom?: TurnForkOrigin
+
+	/** Present on a child session: the session that delegated it. */
+	parentSessionId?: SessionId
 
-	parentRunId?: RunId
+	/** Present on a child session: the parent turn whose tool call spawned it. */
+	parentTurnId?: TurnId
 
 	depth?: number
 
@@ -665,10 +698,10 @@ export interface QueryParams {
 	taskScheduler?: import('../../types/agent/scheduler.js').TaskScheduler
 
 	/**
-	 * Text queued for this run since its last turn, drained at the boundary.
+	 * Text queued for this turn since its last turn, drained at the boundary.
 	 *
 	 * A callback because the queue belongs to whoever accepts the messages,
-	 * and an array captured here would be whatever was queued before the run
+	 * and an array captured here would be whatever was queued before the turn
 	 * started. See `BaseAgentConfig.inboundMessages` for what it closes.
 	 */
 	inboundMessages?: () => import('../../types/message/index.js').Message[]
@@ -681,7 +714,7 @@ export interface QueryParams {
 	waitForInbound?: (signal: AbortSignal) => Promise<void>
 
 	/**
-	 * Live project policy for this run. Unlike `inboundMessages`, snapshot
+	 * Live project policy for this turn. Unlike `inboundMessages`, snapshot
 	 * replacement is durable state and never implies another model turn.
 	 */
 	projectInstructionContext?: ProjectInstructionContext
@@ -689,9 +722,9 @@ export interface QueryParams {
 	/**
 	 * Where this conversation's durable state lives.
 	 *
-	 * Supplies the permission mode when `runConfig.permissionMode` names
+	 * Supplies the permission mode when `turnConfig.permissionMode` names
 	 * none, and receives the flip when a plan is approved. Absent is the
-	 * ordinary case: a run with no topic state behaves exactly as it did.
+	 * ordinary case: a turn with no topic state behaves exactly as it did.
 	 */
 	topicStateStore?: import('../../store/topic/state.js').TopicStateStore
 
@@ -716,15 +749,15 @@ export interface QueryParams {
 	approvalPolicyName?: string
 
 	/**
-	 * Receive this run's approval-policy box, so it can be swapped mid-run.
+	 * Receive this turn's approval-policy box, so it can be swapped mid-run.
 	 *
 	 * The box is built HERE rather than passed in, unlike
 	 * {@link permissionModeRef}, because changing the policy emits a durable
-	 * event and only the run holds the emitter. A host that constructed its
+	 * event and only the turn holds the emitter. A host that constructed its
 	 * own box would be able to change the policy without recording it, which
 	 * is the one thing this must not allow.
 	 */
-	onApprovalPolicy?: (policy: import('../../types/hitl/policy.js').RunApprovalPolicy) => void
+	onApprovalPolicy?: (policy: import('../../types/hitl/policy.js').SessionApprovalPolicy) => void
 
 	/**
 	 * Where a worker completion goes when no tool call is waiting for it.
@@ -747,11 +780,11 @@ export interface QueryParams {
 	compactionConfig?: CompactionConfig
 
 	/**
-	 * Where what the run learned is written when it ends: its decisions,
+	 * Where what the turn learned is written when it ends: its decisions,
 	 * discoveries and failures, as one entry tagged `learning`. Episodic
-	 * memory (the working state) dies with the run; this is the bridge to
-	 * the semantic store a later run searches. Absent means nothing is
-	 * written. A store that fails is logged and never fails the run.
+	 * memory (the working state) dies with the turn; this is the bridge to
+	 * the semantic store a later turn searches. Absent means nothing is
+	 * written. A store that fails is logged and never fails the turn.
 	 */
 	consolidateInto?: MemoryStore
 
@@ -759,13 +792,13 @@ export interface QueryParams {
 	 * Optional neutral working-memory seam. When set, the iteration loop
 	 * re-renders the provider's string into a single pinned leading system
 	 * message every turn (the primacy-edge, compaction-preserved slot).
-	 * Absent ⇒ `refreshWorkingMemory` early-returns and the run path is
+	 * Absent ⇒ `refreshWorkingMemory` early-returns and the turn path is
 	 * byte-identical.
 	 */
 	workingMemoryProvider?: WorkingMemoryProvider
 
 	/**
-	 * Replace context reduction for this run.
+	 * Replace context reduction for this turn.
 	 *
 	 * Outranks `compactionConfig.strategy`, and the built-in structured pass
 	 * does not also run: two mechanisms editing one history in the same pass
@@ -783,10 +816,10 @@ export interface QueryParams {
 	sandboxProvider?: SandboxProvider
 
 	/**
-	 * Maximum time the run waits for sandbox teardown, in milliseconds.
+	 * Maximum time the turn waits for sandbox teardown, in milliseconds.
 	 *
 	 * Defaults to 30 seconds. A fresh private signal is passed to `destroy()`;
-	 * the run also races the returned promise so an implementation that ignores
+	 * the turn also races the returned promise so an implementation that ignores
 	 * cancellation cannot pin `drainQuery()`. Set `0` to retain an unbounded
 	 * teardown wait.
 	 */
@@ -803,7 +836,7 @@ export interface QueryParams {
 	 * explicitly (tool surfaces stripped from prompt + request;
 	 * attachments left unmapped by the driver). The same policy is checked
 	 * immediately before every request for image/document blocks produced by a
-	 * tool, because those do not exist at run setup. `true`: throw instead of
+	 * tool, because those do not exist at turn setup. `true`: throw instead of
 	 * degrading.
 	 */
 	strictCapabilities?: boolean
@@ -815,7 +848,7 @@ export interface QueryParams {
  * The plan re-appends the exact assistant with real/denied/recovered results.
  * Generic history repair must not synthesize a competing result first. Any
  * partial results for that turn are removed too; the executor reconstructs
- * them from the durable run transcript through `recoveredResults`.
+ * them from the durable turn transcript through `recoveredResults`.
  */
 function withoutOwnedResumeTurn(
 	messages: readonly Message[],
@@ -886,10 +919,10 @@ function withOwnedResumeOutcomes(
 	]
 }
 
-export async function* query(params: QueryParams): AsyncGenerator<RunEvent, Run> {
-	const prepared = await prepareRun(params)
+export async function* query(params: QueryParams): AsyncGenerator<SessionEvent, Turn> {
+	const prepared = await prepareTurn(params)
 	const {
-		runConfig,
+		turnConfig,
 		budget,
 		log,
 		ctx,
@@ -902,6 +935,7 @@ export async function* query(params: QueryParams): AsyncGenerator<RunEvent, Run>
 		executeUserInterruptHooks,
 		pendingHistoryRepairs,
 		initialMessages,
+		historyIds,
 		queuedForThisRun,
 		selectedResumeState,
 		attachmentResolutionCancelled,
@@ -912,1465 +946,1505 @@ export async function* query(params: QueryParams): AsyncGenerator<RunEvent, Run>
 	} = prepared
 
 	if (attachmentResolutionCancelled) {
-		return yield* settlePreStartCancellation(params, prepared)
+		try {
+			return yield* settlePreStartCancellation(params, prepared)
+		} finally {
+			await ctx.recorder.release()
+		}
 	}
 
-	const unsubscribeTaskStore = params.taskStore
-		? eventTranslator.wireTaskStore(params.taskStore, ctx.runId)
-		: undefined
-
-	if (params.taskStore) {
-		const taskTools = buildTaskTools(params.taskStore, ctx.runId)
-		const overrides = params.runtimeToolOverrides
-		for (const tool of taskTools) {
-			const override = overrides?.[tool.name]
-			if (override === 'disabled') continue
-			params.tools.register(tool, override ?? 'deferred')
+	// Everything from here to the turn body can refuse (a capability the
+	// provider lacks, a sandbox mode it does not offer); the lease the
+	// prelude took is given back on the way out.
+	let unsubscribeChildSessions: (() => void) | undefined
+	try {
+		const unsubscribeTaskStore = params.taskStore
+			? eventTranslator.wireTaskStore(params.taskStore, ctx.sessionId)
+			: undefined
+		// The children this turn delegates to are recorded in this session's
+		// log: `child_session_spawned` when one starts, `child_session_ended`
+		// (read from the child's own log) when it goes idle. That is what lets
+		// the parent list and replay its children once the process is gone.
+		unsubscribeChildSessions = taskScheduler?.onChildSessionEvent?.((event) => {
+			// Looked up now, while the manager still holds the child's spawn record.
+			const childLog =
+				event.type === 'child_session_idled' ? childSessionLog(event.childSessionId) : undefined
+			void ctx.recorder.recordChildSessionEvent(event, childLog).catch((err: unknown) => {
+				ctx.log.warn('A child session record could not be appended', {
+					[NAMZU.TURN_ID]: ctx.turnId,
+					'namzu.child_session.id': event.childSessionId,
+					'exception.message': err instanceof Error ? err.message : String(err),
+				})
+			})
+		})
+
+		if (params.taskStore) {
+			const taskTools = buildTaskTools(params.taskStore, {
+				sessionId: ctx.sessionId,
+				turnId: ctx.turnId,
+				// A resumed turn began when its `turn_started` was recorded, maybe
+				// in another process: tasks it closed before the pause are still
+				// "closed in this turn".
+				turnStartedAt:
+					(params.resumeFromCheckpoint
+						? await recordedTurnStart(ctx.recorder.log, ctx.turnId)
+						: undefined) ?? ctx.recorder.getTurn().startedAt,
+			})
+			const overrides = params.runtimeToolOverrides
+			for (const tool of taskTools) {
+				const override = overrides?.[tool.name]
+				if (override === 'disabled') continue
+				params.tools.register(tool, override ?? 'deferred')
+			}
 		}
-	}
 
-	if (!params.tools.has(SearchToolsTool.name)) {
-		const hasDeferred = params.tools
-			.listNames()
-			.some((n) => params.tools.getAvailability(n) === 'deferred')
-		if (hasDeferred) {
-			params.tools.register(SearchToolsTool)
+		if (!params.tools.has(SearchToolsTool.name)) {
+			const hasDeferred = params.tools
+				.listNames()
+				.some((n) => params.tools.getAvailability(n) === 'deferred')
+			if (hasDeferred) {
+				params.tools.register(SearchToolsTool)
+			}
 		}
-	}
 
-	// Registered HERE, before the first turn, not when the model is nearly
-	// done. Tools render at prefix position 0, so injecting one late would
-	// invalidate the whole prompt cache for the rest of the run — the same
-	// reason the forced-final turn keeps its tools array and uses
-	// `toolChoice: 'none'` instead of dropping it.
-	if (
-		params.structuredOutput &&
-		params.structuredOutput.mode !== 'native' &&
-		!params.tools.has(STRUCTURED_OUTPUT_TOOL_NAME)
-	) {
-		params.tools.register(createStructuredOutputTool(params.structuredOutput.schema))
-	}
+		// Registered HERE, before the first turn, not when the model is nearly
+		// done. Tools render at prefix position 0, so injecting one late would
+		// invalidate the whole prompt cache for the rest of the turn — the same
+		// reason the forced-final turn keeps its tools array and uses
+		// `toolChoice: 'none'` instead of dropping it.
+		if (
+			params.structuredOutput &&
+			params.structuredOutput.mode !== 'native' &&
+			!params.tools.has(STRUCTURED_OUTPUT_TOOL_NAME)
+		) {
+			params.tools.register(createStructuredOutputTool(params.structuredOutput.schema))
+		}
 
-	// ─── Provider capability negotiation (before tooling bootstrap) ────────
-	// Compare what the request asks for with what the DRIVER declared it
-	// does. Undeclared capabilities resolve permissively (today's behavior
-	// for third-party providers); declared gaps degrade loudly instead of
-	// silently.
-	const capabilities = resolveProviderCapabilities(params.provider)
-	const registeredToolCount = params.tools.listNames().length
-	const stripToolSurfaces = !capabilities.supportsTools && registeredToolCount > 0
-	// Counted separately because they are separate wire shapes: a driver can
-	// map images and drop documents, and a vision warning would send the
-	// reader looking at the wrong half.
-	const carries = (m: (typeof params.messages)[number], kind: 'image' | 'document') =>
-		m.role === 'user' && (m.attachments ?? []).some((a) => (a.type ?? 'image') === kind)
-
-	const attachmentMessageCount = capabilities.supportsVision
-		? 0
-		: params.messages.filter((m) => carries(m, 'image')).length
-	const documentMessageCount = capabilities.supportsDocuments
-		? 0
-		: params.messages.filter((m) => carries(m, 'document')).length
-
-	if (stripToolSurfaces) {
-		const message = `Provider '${params.provider.id}' declares supportsTools: false but ${registeredToolCount} tool(s) are registered — stripping all tool surfaces from the prompt and request so the model is never told about tools it cannot call. Pass strictCapabilities: true to fail instead, or use a tools-capable provider.`
-		if (params.strictCapabilities) {
-			throw new NamzuError({
-				code: 'capability_unavailable',
-				message,
-				details: {
-					providerId: params.provider.id,
-					capability: 'tools',
-					registeredToolCount,
-				},
+		// ─── Provider capability negotiation (before tooling bootstrap) ────────
+		// Compare what the request asks for with what the DRIVER declared it
+		// does. Undeclared capabilities resolve permissively (today's behavior
+		// for third-party providers); declared gaps degrade loudly instead of
+		// silently.
+		const capabilities = resolveProviderCapabilities(params.provider)
+		const registeredToolCount = params.tools.listNames().length
+		const stripToolSurfaces = !capabilities.supportsTools && registeredToolCount > 0
+		// Counted separately because they are separate wire shapes: a driver can
+		// map images and drop documents, and a vision warning would send the
+		// reader looking at the wrong half.
+		const carries = (m: (typeof params.messages)[number], kind: 'image' | 'document') =>
+			m.role === 'user' && (m.attachments ?? []).some((a) => (a.type ?? 'image') === kind)
+
+		const attachmentMessageCount = capabilities.supportsVision
+			? 0
+			: params.messages.filter((m) => carries(m, 'image')).length
+		const documentMessageCount = capabilities.supportsDocuments
+			? 0
+			: params.messages.filter((m) => carries(m, 'document')).length
+
+		if (stripToolSurfaces) {
+			const message = `Provider '${params.provider.id}' declares supportsTools: false but ${registeredToolCount} tool(s) are registered — stripping all tool surfaces from the prompt and request so the model is never told about tools it cannot call. Pass strictCapabilities: true to fail instead, or use a tools-capable provider.`
+			if (params.strictCapabilities) {
+				throw new NamzuError({
+					code: 'capability_unavailable',
+					message,
+					details: {
+						providerId: params.provider.id,
+						capability: 'tools',
+						registeredToolCount,
+					},
+				})
+			}
+			ctx.log.warn('Capability mismatch: the provider declares no tool support', {
+				'namzu.capability.detail': message,
+				[GENAI.SYSTEM]: params.provider.id,
+				'namzu.runtime.registered_tool_count': registeredToolCount,
 			})
 		}
-		ctx.log.warn('Capability mismatch: the provider declares no tool support', {
-			'namzu.capability.detail': message,
-			[GENAI.SYSTEM]: params.provider.id,
-			'namzu.runtime.registered_tool_count': registeredToolCount,
-		})
-	}
 
-	if (attachmentMessageCount > 0) {
-		const message = `Provider '${params.provider.id}' declares supportsVision: false but ${attachmentMessageCount} user message(s) carry image attachments — the driver will not map them, so the model never sees the images. Pass strictCapabilities: true to fail instead, or use a vision-capable provider.`
-		if (params.strictCapabilities) {
-			throw new NamzuError({
-				code: 'capability_unavailable',
-				message,
-				details: {
-					providerId: params.provider.id,
-					capability: 'vision',
-					attachmentMessageCount,
-				},
+		if (attachmentMessageCount > 0) {
+			const message = `Provider '${params.provider.id}' declares supportsVision: false but ${attachmentMessageCount} user message(s) carry image attachments — the driver will not map them, so the model never sees the images. Pass strictCapabilities: true to fail instead, or use a vision-capable provider.`
+			if (params.strictCapabilities) {
+				throw new NamzuError({
+					code: 'capability_unavailable',
+					message,
+					details: {
+						providerId: params.provider.id,
+						capability: 'vision',
+						attachmentMessageCount,
+					},
+				})
+			}
+			ctx.log.warn('Capability mismatch: the provider declares no vision support', {
+				'namzu.capability.detail': message,
+				[GENAI.SYSTEM]: params.provider.id,
+				'namzu.runtime.attachment_message_count': attachmentMessageCount,
 			})
 		}
-		ctx.log.warn('Capability mismatch: the provider declares no vision support', {
-			'namzu.capability.detail': message,
-			[GENAI.SYSTEM]: params.provider.id,
-			'namzu.runtime.attachment_message_count': attachmentMessageCount,
-		})
-	}
 
-	if (documentMessageCount > 0) {
-		const message = `Provider '${params.provider.id}' declares supportsDocuments: false but ${documentMessageCount} user message(s) carry document attachments — the driver will not map them, so the model never sees the documents. Pass strictCapabilities: true to fail instead, or use a document-capable provider.`
-		if (params.strictCapabilities) {
-			throw new NamzuError({
-				code: 'capability_unavailable',
-				message,
-				details: {
-					providerId: params.provider.id,
-					capability: 'documents',
-					documentMessageCount,
-				},
+		if (documentMessageCount > 0) {
+			const message = `Provider '${params.provider.id}' declares supportsDocuments: false but ${documentMessageCount} user message(s) carry document attachments — the driver will not map them, so the model never sees the documents. Pass strictCapabilities: true to fail instead, or use a document-capable provider.`
+			if (params.strictCapabilities) {
+				throw new NamzuError({
+					code: 'capability_unavailable',
+					message,
+					details: {
+						providerId: params.provider.id,
+						capability: 'documents',
+						documentMessageCount,
+					},
+				})
+			}
+			ctx.log.warn('Capability mismatch: the provider declares no document support', {
+				'namzu.capability.detail': message,
+				[GENAI.SYSTEM]: params.provider.id,
+				'namzu.runtime.document_message_count': documentMessageCount,
 			})
 		}
-		ctx.log.warn('Capability mismatch: the provider declares no document support', {
-			'namzu.capability.detail': message,
-			[GENAI.SYSTEM]: params.provider.id,
-			'namzu.runtime.document_message_count': documentMessageCount,
-		})
-	}
-
-	// Denied names are subtracted LAST, after the allow-list is resolved
-	// against the registry — so a deny reaches a run that named no
-	// allow-list at all, which is the ordinary case for a delegated child.
-	// Applied to `effectiveAllowedTools`, which `query()` binds to both the
-	// request tool list AND the `ToolExecutor`: narrowing only the request
-	// shows the model fewer tools and lets it call any of them by name.
-	const allowedBeforeDenial = stripToolSurfaces
-		? []
-		: withDeferredDiscoveryTool(params.tools, params.allowedTools)
-	const denied = new Set(params.deniedTools ?? [])
-	const effectiveAllowedTools: string[] | undefined =
-		denied.size === 0
-			? allowedBeforeDenial
-			: // An absent allow-list means "every registered tool", so a deny
-				// with no allow-list has to be resolved against the registry
-				// here — otherwise the subtraction would be from an empty list
-				// and would deny nothing, which is the shape a delegated child
-				// arrives in.
-				[...(allowedBeforeDenial ?? params.tools.listNames())].filter((name) => !denied.has(name))
-
-	// The two halves of a durable pause, owned by the RUN when the host
-	// does not own them.
-	//
-	// `SupervisorAgent` builds both before the run exists, because the
-	// tools it builds close over them, and it passes them in. Nothing else
-	// could: neither type is exported from `public-runtime.ts`, so a host
-	// on `ReactiveAgent`, `drainQuery` or `resumeRun` had no way to supply
-	// either — and `ToolContext.requestPause`, which every tool author is
-	// handed, silently wrote no checkpoint and could receive no answer on
-	// those surfaces. Which agent class the host happened to pick is not
-	// visible at the call site, so the degradation was invisible too.
-	//
-	// A run-local pair is enough for the general seam because `query()`
-	// builds its `createToolPause` itself, below, and can hand it the
-	// run's own. Pinned by the "a pause is durable on any surface" cases
-	// in `__tests__/tool-pause-resume.test.ts`.
-	const questionParks = params.questionParks ?? new QuestionParkBinding()
-	const pendingAnswers = params.pendingAnswers ?? new PendingAnswers()
-
-	// One gate instance owns both model-issued calls and calls dispatched by
-	// another tool. Constructing it only inside the iteration review left the
-	// nested registry path outside the operator's policy entirely.
-	const gateConfig = params.authorizationGate
-	const verificationGate = gateConfig?.enabled
-		? new AuthorizationGate(gateConfig, ctx.log)
-		: undefined
-
-	// The store is initialized when the async query starts, after tooling is
-	// composed. Resolve lazily at execution time: capturing getRunDir() here
-	// permanently disabled retention for fresh disk-backed invocations.
-	const toolOutputDir = () => {
-		const runDir = ctx.runMgr.getRunDir()
-		return runDir ? join(runDir, TOOL_OUTPUT_DIR_NAME) : undefined
-	}
-	// The same invocation-owned capability serves tools and optional preparation.
-	// Local cancellation cannot override the run's cancellation or settled state.
-	const captureRunEvidence = async (maxReadBytes?: number, signal?: AbortSignal) => {
-		const combined = signal
-			? AbortSignal.any([ctx.abortController.signal, signal])
-			: ctx.abortController.signal
-		combined.throwIfAborted()
-		const source = await eventTranslator.captureRunEvidence(maxReadBytes, combined)
-		combined.throwIfAborted()
-		return source
-	}
 
-	// Whose jobs this run speaks for: its own by default, the session's when
-	// the host said so. Resolved before the tools are built, because the
-	// wait-intent recorder below is bound into them.
-	const jobOwner = params.backgroundJobOwner ?? ctx.runId
-	// What the model reads when a job ends. Built up here, ahead of the
-	// subscription that fills it below, because the wait-intent recorder needs
-	// to ask whether its text has been read yet.
-	const jobNotices = params.backgroundJobs ? new SteeringBinding() : undefined
-	// Jobs the model told `wait_for_job` it is waiting on, which is the only
-	// thing that can hold this run open for a job. Built only where there is a
-	// registry, so a host with no background mode carries no recorder and the
-	// bound ref has no `markAwaited` to offer.
-	const awaitedJobs = params.backgroundJobs
-		? new AwaitedJobs(params.backgroundJobs, jobOwner, () => jobNotices?.pending ?? false)
-		: undefined
-	awaitedJobs?.attach()
-
-	const toolExecutor = ToolingBootstrap.init(
-		{
-			tools: params.tools,
-			runId: ctx.runId,
-			workingDirectory: ctx.cwd,
-			...(params.additionalDirectories?.length
-				? { additionalDirectories: params.additionalDirectories }
-				: {}),
-			permissionMode: () => ctx.permissionMode.current,
-			env: runConfig.env ?? {},
-			abortSignal: ctx.abortController.signal,
-			allowedTools: effectiveAllowedTools,
-			invocationState: params.invocationState,
-			pluginManager: params.pluginManager,
-			...(params.backgroundJobs ? { backgroundJobs: params.backgroundJobs } : {}),
-			...(params.backgroundJobOwner ? { backgroundJobOwner: params.backgroundJobOwner } : {}),
-			...(awaitedJobs ? { onJobAwaited: (id: string) => awaitedJobs.expect(id) } : {}),
-			// The `skill` tool's registry. Threaded from the run rather than
-			// held by the tool, because a tool that reached for a module-level
-			// registry would answer about whatever the last run configured.
-			...(params.skillRegistry ? { skills: params.skillRegistry } : {}),
-			...(params.web ? { web: params.web } : {}),
-			...(params.fileReadTracker ? { fileReadTracker: params.fileReadTracker } : {}),
-			...(params.toolTimeoutMs !== undefined ? { toolTimeoutMs: params.toolTimeoutMs } : {}),
-			...(params.toolRetryBackoff !== undefined
-				? { toolRetryBackoff: params.toolRetryBackoff }
-				: {}),
-			...(params.maxToolCalls !== undefined
-				? {
-						maxToolCalls: params.maxToolCalls,
-						readToolCallBudgetEvents: () => eventTranslator.readEvents({ integrity: 'strict' }),
-					}
-				: {}),
-			...(params.maxToolConcurrency !== undefined
-				? { maxToolConcurrency: params.maxToolConcurrency }
-				: {}),
-			...(params.maxToolOutputChars !== undefined
-				? { maxToolOutputChars: params.maxToolOutputChars }
-				: {}),
-			...(params.toolResultGuardrails !== undefined
-				? { toolResultGuardrails: params.toolResultGuardrails }
-				: {}),
-			...(params.retainedToolPreviewChars !== undefined
-				? { retainedToolPreviewChars: params.retainedToolPreviewChars }
-				: {}),
-			...(params.maxToolContentBytes !== undefined
-				? { maxToolContentBytes: params.maxToolContentBytes }
-				: {}),
-			// Overflow lands beside the run's other artifacts, so it is
-			// cleaned up with the run and reachable by the model's own
-			// `read`/`grep` without a new affordance.
-			toolOutputDir,
-			captureRunEvidence,
-			...(params.repairToolCall ? { repairToolCall: params.repairToolCall } : {}),
-			...(verificationGate ? { authorizationGate: verificationGate } : {}),
-			recordAudit: (input) => ctx.runMgr.recordAudit(input),
-			// The durable pause, reachable from any tool rather than from the
-			// four kernel-owned points that used to own it. Built here from
-			// the machinery the run already holds; the recorder binds a few
-			// lines below, and until it does a pause is in-process only —
-			// the same degradation the built-in question tool has.
-			toolPause: (toolUseId) =>
-				createToolPause({
-					runId: ctx.runId,
-					toolUseId,
-					parkHandler: (request) => approvalPolicy.current.handler(request),
-					recorder: questionParks,
-					pendingAnswers,
-				}),
-		},
-		ctx.activityStore,
-		eventTranslator.emitEvent,
-		ctx.log,
-	)
+		// Denied names are subtracted LAST, after the allow-list is resolved
+		// against the registry — so a deny reaches a turn that named no
+		// allow-list at all, which is the ordinary case for a delegated child.
+		// Applied to `effectiveAllowedTools`, which `query()` binds to both the
+		// request tool list AND the `ToolExecutor`: narrowing only the request
+		// shows the model fewer tools and lets it call any of them by name.
+		const allowedBeforeDenial = stripToolSurfaces
+			? []
+			: withDeferredDiscoveryTool(params.tools, params.allowedTools)
+		const denied = new Set(params.deniedTools ?? [])
+		const effectiveAllowedTools: string[] | undefined =
+			denied.size === 0
+				? allowedBeforeDenial
+				: // An absent allow-list means "every registered tool", so a deny
+					// with no allow-list has to be resolved against the registry
+					// here — otherwise the subtraction would be from an empty list
+					// and would deny nothing, which is the shape a delegated child
+					// arrives in.
+					[...(allowedBeforeDenial ?? params.tools.listNames())].filter((name) => !denied.has(name))
+
+		// The two halves of a durable pause, owned by the TURN when the host
+		// does not own them.
+		//
+		// `SupervisorAgent` builds both before the turn exists, because the
+		// tools it builds close over them, and it passes them in. Nothing else
+		// could: neither type is exported from `public-runtime.ts`, so a host
+		// on `ReactiveAgent`, `drainQuery` or `resumeSession` had no way to supply
+		// either — and `ToolContext.requestPause`, which every tool author is
+		// handed, silently wrote no checkpoint and could receive no answer on
+		// those surfaces. Which agent class the host happened to pick is not
+		// visible at the call site, so the degradation was invisible too.
+		//
+		// A run-local pair is enough for the general seam because `query()`
+		// builds its `createToolPause` itself, below, and can hand it the
+		// run's own. Pinned by the "a pause is durable on any surface" cases
+		// in `__tests__/tool-pause-resume.test.ts`.
+		const questionParks = params.questionParks ?? new QuestionParkBinding()
+		const pendingAnswers = params.pendingAnswers ?? new PendingAnswers()
+
+		// One gate instance owns both model-issued calls and calls dispatched by
+		// another tool. Constructing it only inside the iteration review left the
+		// nested registry path outside the operator's policy entirely.
+		const gateConfig = params.authorizationGate
+		const verificationGate = gateConfig?.enabled
+			? new AuthorizationGate(gateConfig, ctx.log)
+			: undefined
+
+		// Oversized tool output is spilled beside the session log
+		// (`<session-id>/tool-results/`). An in-memory session has nowhere to
+		// spill it, so the ordinary output budget applies.
+		const toolOutputDir = () => ctx.toolResultsDir
+		// The same invocation-owned capability serves tools and optional preparation.
+		// Local cancellation cannot override the turn's cancellation or settled state.
+		const captureSessionEvidence = async (maxReadBytes?: number, signal?: AbortSignal) => {
+			const combined = signal
+				? AbortSignal.any([ctx.abortController.signal, signal])
+				: ctx.abortController.signal
+			combined.throwIfAborted()
+			const source = await eventTranslator.captureSessionEvidence(maxReadBytes, combined)
+			combined.throwIfAborted()
+			return source
+		}
 
-	// A background job's exit reaches the model as a notice on its next tool
-	// result, and the host as an event — without either polling. Subscribed
-	// for the owner the run's jobs are bound to, so a session-owned job that
-	// ends during this run is reported here too.
-	const unsubscribeJobExits = params.backgroundJobs?.onExit((job) => {
-		if (job.owner !== jobOwner) return
-		const outcome =
-			job.status === 'killed'
-				? 'was stopped'
-				: `exited with code ${job.exitCode ?? 'unknown'}${job.signal ? ` (${job.signal})` : ''}`
-		jobNotices?.steer(
-			`Background job ${job.id} (${job.command}) ${outcome}. Read its output with the job tool if you need it.`,
+		// Whose jobs this turn speaks for: its own by default, the session's when
+		// the host said so. Resolved before the tools are built, because the
+		// wait-intent recorder below is bound into them.
+		const jobOwner = params.backgroundJobOwner ?? ctx.turnId
+		// What the model reads when a job ends. Built up here, ahead of the
+		// subscription that fills it below, because the wait-intent recorder needs
+		// to ask whether its text has been read yet.
+		const jobNotices = params.backgroundJobs ? new SteeringBinding() : undefined
+		// Jobs the model told `wait_for_job` it is waiting on, which is the only
+		// thing that can hold this turn open for a job. Built only where there is a
+		// registry, so a host with no background mode carries no recorder and the
+		// bound ref has no `markAwaited` to offer.
+		const awaitedJobs = params.backgroundJobs
+			? new AwaitedJobs(params.backgroundJobs, jobOwner, () => jobNotices?.pending ?? false)
+			: undefined
+		awaitedJobs?.attach()
+
+		const toolExecutor = ToolingBootstrap.init(
+			{
+				tools: params.tools,
+				sessionId: ctx.sessionId,
+				turnId: ctx.turnId,
+				workingDirectory: ctx.cwd,
+				...(params.additionalDirectories?.length
+					? { additionalDirectories: params.additionalDirectories }
+					: {}),
+				...(params.outsideRootAccess ? { outsideRootAccess: params.outsideRootAccess } : {}),
+				...(params.sandboxEscape ? { sandboxEscape: params.sandboxEscape } : {}),
+				permissionMode: () => ctx.permissionMode.current,
+				env: turnConfig.env ?? {},
+				abortSignal: ctx.abortController.signal,
+				allowedTools: effectiveAllowedTools,
+				invocationState: params.invocationState,
+				pluginManager: params.pluginManager,
+				...(params.backgroundJobs ? { backgroundJobs: params.backgroundJobs } : {}),
+				...(params.backgroundJobOwner ? { backgroundJobOwner: params.backgroundJobOwner } : {}),
+				...(awaitedJobs ? { onJobAwaited: (id: string) => awaitedJobs.expect(id) } : {}),
+				// The `skill` tool's registry. Threaded from the turn rather than
+				// held by the tool, because a tool that reached for a module-level
+				// registry would answer about whatever the last turn configured.
+				...(params.skillRegistry ? { skills: params.skillRegistry } : {}),
+				...(params.web ? { web: params.web } : {}),
+				...(params.fileReadTracker ? { fileReadTracker: params.fileReadTracker } : {}),
+				...(params.toolTimeoutMs !== undefined ? { toolTimeoutMs: params.toolTimeoutMs } : {}),
+				...(params.toolRetryBackoff !== undefined
+					? { toolRetryBackoff: params.toolRetryBackoff }
+					: {}),
+				...(params.maxToolCalls !== undefined
+					? {
+							maxToolCalls: params.maxToolCalls,
+							readToolCallBudgetRecords: () => eventTranslator.readRecords({ mode: 'strict' }),
+						}
+					: {}),
+				...(params.maxToolConcurrency !== undefined
+					? { maxToolConcurrency: params.maxToolConcurrency }
+					: {}),
+				...(params.maxToolOutputChars !== undefined
+					? { maxToolOutputChars: params.maxToolOutputChars }
+					: {}),
+				...(params.toolResultGuardrails !== undefined
+					? { toolResultGuardrails: params.toolResultGuardrails }
+					: {}),
+				...(params.retainedToolPreviewChars !== undefined
+					? { retainedToolPreviewChars: params.retainedToolPreviewChars }
+					: {}),
+				...(params.maxToolContentBytes !== undefined
+					? { maxToolContentBytes: params.maxToolContentBytes }
+					: {}),
+				// Overflow lands beside the turn's other artifacts, so it is
+				// cleaned up with the turn and reachable by the model's own
+				// `read`/`grep` without a new affordance.
+				toolOutputDir,
+				captureSessionEvidence,
+				...(params.repairToolCall ? { repairToolCall: params.repairToolCall } : {}),
+				...(verificationGate ? { authorizationGate: verificationGate } : {}),
+				recordAudit: (input) => ctx.recorder.recordAudit(input),
+				// The durable pause, reachable from any tool rather than from the
+				// four kernel-owned points that used to own it. Built here from
+				// the machinery the turn already holds; the recorder binds a few
+				// lines below, and until it does a pause is in-process only —
+				// the same degradation the built-in question tool has.
+				toolPause: (toolUseId) =>
+					createToolPause({
+						sessionId: ctx.sessionId,
+						turnId: ctx.turnId,
+						toolUseId,
+						parkHandler: (request) => approvalPolicy.current.handler(request),
+						recorder: questionParks,
+						pendingAnswers,
+					}),
+			},
+			ctx.activityStore,
+			eventTranslator.emitEvent,
+			ctx.log,
 		)
-		void eventTranslator
-			.emitEvent({
-				type: 'background_job_exited',
-				runId: ctx.runId,
-				jobId: job.id,
-				command: job.command,
-				status: job.status === 'killed' ? 'killed' : 'exited',
-				...(job.exitCode !== undefined ? { exitCode: job.exitCode } : {}),
-				...(job.signal ? { signal: job.signal } : {}),
-			})
-			.catch(() => {})
-	})
-	let workingStateManager: WorkingStateManager | undefined
-	// Normalised once, defaults applied. A host that passes a partial
-	// object — a strategy and a window, nothing else — used to reach the
-	// phase with `triggerThreshold` undefined, and `usage < undefined` is
-	// false: the pass ran on every iteration and the salience strategy fell
-	// through to the stale-result clearing it exists to replace. The eval
-	// suite that passed exactly that object found it.
-	const compactionConfig = params.compactionConfig
-		? CompactionConfigSchema.parse(params.compactionConfig)
-		: undefined
-	if (compactionConfig && compactionConfig.strategy !== 'disabled') {
-		workingStateManager = new WorkingStateManager(compactionConfig)
-		toolExecutor.setWorkingStateManager(workingStateManager)
-	}
 
-	// The run's own registry when the host passed none, so a hook's
-	// annotation has somewhere to land.
-	const promptContributions = params.promptContributions ?? new PromptContributionRegistry()
-
-	const promptBuilder = new PromptBuilder({
-		systemPrompt: params.systemPrompt,
-		persona: params.persona,
-		skills: params.skills,
-		basePrompt: params.basePrompt,
-		tools: params.tools,
-		allowedTools: effectiveAllowedTools,
-		runtimeContext: params.runtimeContext,
-		contributions: promptContributions,
-	})
+		// A background job's exit reaches the model as a notice on its next tool
+		// result, and the host as an event — without either polling. Subscribed
+		// for the owner the turn's jobs are bound to, so a session-owned job that
+		// ends during this turn is reported here too.
+		const unsubscribeJobExits = params.backgroundJobs?.onExit((job) => {
+			if (job.owner !== jobOwner) return
+			const outcome =
+				job.status === 'killed'
+					? 'was stopped'
+					: `exited with code ${job.exitCode ?? 'unknown'}${job.signal ? ` (${job.signal})` : ''}`
+			jobNotices?.steer(
+				`Background job ${job.id} (${job.command}) ${outcome}. Read its output with the job tool if you need it.`,
+			)
+			void eventTranslator
+				.emitEvent({
+					type: 'background_job_exited',
+					turnId: ctx.turnId,
+					jobId: job.id,
+					command: job.command,
+					status: job.status === 'killed' ? 'killed' : 'exited',
+					...(job.exitCode !== undefined ? { exitCode: job.exitCode } : {}),
+					...(job.signal ? { signal: job.signal } : {}),
+				})
+				.catch(() => {})
+		})
+		let workingStateManager: WorkingStateManager | undefined
+		// Normalised once, defaults applied. A host that passes a partial
+		// object — a strategy and a window, nothing else — used to reach the
+		// phase with `triggerThreshold` undefined, and `usage < undefined` is
+		// false: the pass ran on every iteration and the salience strategy fell
+		// through to the stale-result clearing it exists to replace. The eval
+		// suite that passed exactly that object found it.
+		const compactionConfig = params.compactionConfig
+			? CompactionConfigSchema.parse(params.compactionConfig)
+			: undefined
+		if (compactionConfig && compactionConfig.strategy !== 'disabled') {
+			workingStateManager = new WorkingStateManager(compactionConfig)
+			toolExecutor.setWorkingStateManager(workingStateManager)
+		}
 
-	const guard = new GuardCoordinator({
-		tokenBudget: runConfig.tokenBudget,
-		timeoutMs: runConfig.timeoutMs,
-		costLimitUsd: runConfig.costLimitUsd,
-		maxIterations: runConfig.maxIterations,
-	})
+		// The turn's own registry when the host passed none, so a hook's
+		// annotation has somewhere to land.
+		const promptContributions = params.promptContributions ?? new PromptContributionRegistry()
 
-	const checkpointMgr = new CheckpointManager(
-		ctx.runMgr.getCheckpointStore(),
-		ctx.runMgr.getRunScope(),
-	)
+		const promptBuilder = new PromptBuilder({
+			systemPrompt: params.systemPrompt,
+			persona: params.persona,
+			skills: params.skills,
+			basePrompt: params.basePrompt,
+			tools: params.tools,
+			allowedTools: effectiveAllowedTools,
+			runtimeContext: params.runtimeContext,
+			contributions: promptContributions,
+		})
 
-	// Every checkpoint carries compaction's accumulated state, so a run that
-	// comes back in a new process can adopt it (see the restore block).
-	// Without it, compaction's own justification for dropping the prior
-	// `[COMPACTED CONTEXT]` block — that `serializeState` is cumulative —
-	// holds within one process and fails across a resume.
-	if (workingStateManager) {
-		const manager = workingStateManager
-		checkpointMgr.setWorkingStateSource(() => snapshotWorkingState(manager))
-	}
+		const guard = new GuardCoordinator({
+			tokenBudget: turnConfig.tokenBudget,
+			timeoutMs: turnConfig.timeoutMs,
+			costLimitUsd: turnConfig.costLimitUsd,
+			maxIterations: turnConfig.maxIterations,
+		})
 
-	const resultAssembler = new ResultAssembler({
-		runMgr: ctx.runMgr,
-		planManager: ctx.planManager,
-		activityStore: ctx.activityStore,
-		log: ctx.log,
-		emitEvent: eventTranslator.emitEvent,
-		drainPending: () => eventTranslator.drainPending(),
-		// Read at settle time, not now: checkpoints are written per
-		// iteration, so the answer changes as the run proceeds.
-		resumeCheckpointId: () => checkpointMgr.lastCheckpointId,
-		// Read only to recover WHY a cancellation happened. The run loop
-		// already knows THAT it was cancelled; the origin lives on the abort
-		// reason and nothing else carries it this far.
-		signal: ctx.abortController.signal,
-	})
+		const checkpointMgr = new CheckpointManager(ctx.recorder, ctx.storage.checkpoints, {
+			tenantId: ctx.tenantId,
+			projectId: ctx.projectId,
+			sessionId: ctx.sessionId,
+			turnId: ctx.turnId,
+		})
 
-	let advisoryCtx: AdvisoryContext | undefined
-	if (params.advisory && params.advisory.advisors.length > 0) {
-		// Advisors are model calls owned by this run even when they use a
-		// different provider. Sending the raw definitions into the registry
-		// lets a triggered or model-requested consultation bypass both the
-		// finite stream-silence bound and Stop. Bind every advisor provider at
-		// the query boundary, where the effective timeout and run signal are
-		// already known; standalone AdvisoryExecutor callers retain their
-		// explicitly chosen provider/cancellation policy.
-		const boundedAdvisors = params.advisory.advisors.map((advisor) => ({
-			...advisor,
-			provider: withTokenBudget(
-				withStreamIdleTimeout(advisor.provider, {
-					idleTimeoutMs: streamIdleTimeoutMs,
-					log: ctx.log,
-				}),
-				budget,
-			),
-		}))
-		const advisorRegistry = new AdvisorRegistry(boundedAdvisors, params.advisory.defaultAdvisorId)
-		// A budget the runtime cannot measure is refused here rather than
-		// silently ignored for the length of the run.
-		assertBudgetEnforceable(params.advisory)
-		const advisoryExecutor = new AdvisoryExecutor(
-			ctx.log,
-			params.advisory.budget,
-			ctx.abortController.signal,
-		)
-		const triggerEvaluator = new TriggerEvaluator(
-			params.advisory.triggers ?? [],
-			params.advisory.budget,
-		)
-		advisoryCtx = new AdvisoryContext(
-			advisorRegistry,
-			advisoryExecutor,
-			triggerEvaluator,
-			params.advisory.budget,
-		)
+		// Every checkpoint carries compaction's accumulated state, so a turn that
+		// comes back in a new process can adopt it (see the restore block).
+		// Without it, compaction's own justification for dropping the prior
+		// `[COMPACTED CONTEXT]` block — that `serializeState` is cumulative —
+		// holds within one process and fails across a resume.
+		if (workingStateManager) {
+			const manager = workingStateManager
+			checkpointMgr.setWorkingStateSource(() => snapshotWorkingState(manager))
+		}
 
-		// What the run looks like when the MODEL consults an advisor, as
-		// opposed to when a trigger does. The trigger path has always passed
-		// this; the tool path passed an empty context, so an advisor the model
-		// asked for help saw the question and nothing else.
-		//
-		// `includeToolCatalog` and `useCompactedContext` are read here and
-		// nowhere else. Both were declared on `AdvisoryConfig` /
-		// `AdvisorDefinition` and consulted by nothing, so a host who turned
-		// the catalogue off still paid for it in every advisory prompt.
-		const advisoryConfig = params.advisory
-		advisoryCtx.setCallContextProvider(() => {
-			const summary =
-				workingStateManager && advisoryConfig.advisors.some((a) => a.useCompactedContext)
-					? serializeWorkingState(workingStateManager.getState())
-					: undefined
-			return {
-				messages: ctx.runMgr.messages,
-				turn: iterationOrchestrator.getAdvisoryTurnContext(),
-				...(summary !== undefined ? { workingStateSummary: summary } : {}),
-				...(advisoryConfig.includeToolCatalog
-					? { toolCatalog: params.tools.toLLMTools(effectiveAllowedTools) }
-					: {}),
-				iteration: ctx.runMgr.currentIteration,
-			}
+		const resultAssembler = new ResultAssembler({
+			recorder: ctx.recorder,
+			planManager: ctx.planManager,
+			activityStore: ctx.activityStore,
+			log: ctx.log,
+			emitEvent: eventTranslator.emitEvent,
+			drainPending: () => eventTranslator.drainPending(),
+			// Read at settle time, not now: checkpoints are written per
+			// iteration, so the answer changes as the turn proceeds. A turn
+			// that fails before writing one (a 429 on its first request)
+			// commits the state its loop began from, so it pauses like any
+			// later failure rather than failing with nothing to resume.
+			resumeCheckpointId: async () =>
+				checkpointMgr.lastCheckpointId ?? (await checkpointMgr.createAtLoopStart(ctx.recorder))?.id,
+			// Read only to recover WHY a cancellation happened. The turn loop
+			// already knows THAT it was cancelled; the origin lives on the abort
+			// reason and nothing else carries it this far.
+			signal: ctx.abortController.signal,
 		})
 
-		if (params.advisory.enableAgentTool) {
-			const advisoryTools = buildAdvisoryTools({ advisoryCtx })
-			const overrides = params.runtimeToolOverrides
-			for (const tool of advisoryTools) {
-				const override = overrides?.[tool.name]
-				if (override === 'disabled') continue
-				params.tools.register(tool, override ?? 'active')
+		let advisoryCtx: AdvisoryContext | undefined
+		if (params.advisory && params.advisory.advisors.length > 0) {
+			// Advisors are model calls owned by this turn even when they use a
+			// different provider. Sending the raw definitions into the registry
+			// lets a triggered or model-requested consultation bypass both the
+			// finite stream-silence bound and Stop. Bind every advisor provider at
+			// the query boundary, where the effective timeout and run signal are
+			// already known; standalone AdvisoryExecutor callers retain their
+			// explicitly chosen provider/cancellation policy.
+			const boundedAdvisors = params.advisory.advisors.map((advisor) => ({
+				...advisor,
+				provider: withTokenBudget(
+					withStreamIdleTimeout(advisor.provider, {
+						idleTimeoutMs: streamIdleTimeoutMs,
+						log: ctx.log,
+					}),
+					budget,
+				),
+			}))
+			const advisorRegistry = new AdvisorRegistry(boundedAdvisors, params.advisory.defaultAdvisorId)
+			// A budget the runtime cannot measure is refused here rather than
+			// silently ignored for the length of the turn.
+			assertBudgetEnforceable(params.advisory)
+			const advisoryExecutor = new AdvisoryExecutor(
+				ctx.log,
+				params.advisory.budget,
+				ctx.abortController.signal,
+			)
+			const triggerEvaluator = new TriggerEvaluator(
+				params.advisory.triggers ?? [],
+				params.advisory.budget,
+			)
+			advisoryCtx = new AdvisoryContext(
+				advisorRegistry,
+				advisoryExecutor,
+				triggerEvaluator,
+				params.advisory.budget,
+			)
+
+			// What the turn looks like when the MODEL consults an advisor, as
+			// opposed to when a trigger does. The trigger path has always passed
+			// this; the tool path passed an empty context, so an advisor the model
+			// asked for help saw the question and nothing else.
+			//
+			// `includeToolCatalog` and `useCompactedContext` are read here and
+			// nowhere else. Both were declared on `AdvisoryConfig` /
+			// `AdvisorDefinition` and consulted by nothing, so a host who turned
+			// the catalogue off still paid for it in every advisory prompt.
+			const advisoryConfig = params.advisory
+			advisoryCtx.setCallContextProvider(() => {
+				const summary =
+					workingStateManager && advisoryConfig.advisors.some((a) => a.useCompactedContext)
+						? serializeWorkingState(workingStateManager.getState())
+						: undefined
+				return {
+					messages: ctx.recorder.messages,
+					turn: iterationOrchestrator.getAdvisoryTurnContext(),
+					...(summary !== undefined ? { workingStateSummary: summary } : {}),
+					...(advisoryConfig.includeToolCatalog
+						? { toolCatalog: params.tools.toLLMTools(effectiveAllowedTools) }
+						: {}),
+					iteration: ctx.recorder.currentIteration,
+				}
+			})
+
+			if (params.advisory.enableAgentTool) {
+				const advisoryTools = buildAdvisoryTools({ advisoryCtx })
+				const overrides = params.runtimeToolOverrides
+				for (const tool of advisoryTools) {
+					const override = overrides?.[tool.name]
+					if (override === 'disabled') continue
+					params.tools.register(tool, override ?? 'active')
+				}
 			}
 		}
-	}
 
-	const iterationOrchestrator = new IterationOrchestrator({
-		captureRunEvidence,
-		provider: resilientProvider,
-		providerCapabilities: capabilities,
-		strictCapabilities: params.strictCapabilities === true,
-		servingMember: () => serving.current,
-		runConfig,
-		...(params.stopWhen ? { stopWhen: params.stopWhen } : {}),
-		...(params.prepareStep ? { prepareStep: params.prepareStep } : {}),
-		...(params.beforeStep ? { beforeStep: params.beforeStep } : {}),
-		...(params.onStepFinish ? { onStepFinish: params.onStepFinish } : {}),
-		...(params.reviewAnswer ? { reviewAnswer: params.reviewAnswer } : {}),
-		...(params.maxAnswerReviews !== undefined ? { maxAnswerReviews: params.maxAnswerReviews } : {}),
-		...(params.structuredOutput ? { structuredOutput: params.structuredOutput } : {}),
-		...(params.parkRecordDelayMs !== undefined
-			? { parkRecordDelayMs: params.parkRecordDelayMs }
-			: {}),
-		tools: params.tools,
-		allowedTools: effectiveAllowedTools,
-		runMgr: ctx.runMgr,
-		toolExecutor,
-		guard,
-		activityStore: ctx.activityStore,
-		emitEvent: eventTranslator.emitEvent,
-		drainPending: () => eventTranslator.drainPending(),
-		abortController: ctx.abortController,
-		log: ctx.log,
-		// Read through the box on every call, so a swap lands on the next
-		// question rather than the next run.
-		resumeHandler: (request) => approvalPolicy.current.handler(request),
-		takeApprovalPolicyChange: () => approvalPolicy.takeUnannouncedChange(),
-		promptContributions,
-		...(params.steering ? { steering: params.steering } : {}),
-		...(jobNotices ? { jobNotices } : {}),
-		...(awaitedJobs ? { awaitedJobs } : {}),
-		checkpointMgr,
-		planManager: ctx.planManager,
-		taskGateway: taskScheduler,
-		completionInbox: params.completionInbox,
-		taskStore: params.taskStore,
-		// Run-scoped. An approval is a statement about this run's work;
-		// carrying one into a later run would be reuse nobody agreed to.
-		toolGrants: new ToolGrantSet(),
-		// Run-scoped for the same reason. A repeat count carried into a later
-		// run is a claim about work nobody repeated, and a module-level map
-		// would leak exactly that way.
-		...(params.repeatCallAdvisory === false ? {} : { repeatCalls: new RepeatCallTracker() }),
-		compactionConfig,
-		...(params.inboundMessages ? { inboundMessages: params.inboundMessages } : {}),
-		...(params.resumeFromCheckpoint ? { resumedInput: queuedForThisRun } : {}),
-		...(params.waitForInbound ? { waitForInbound: params.waitForInbound } : {}),
-		...(params.projectInstructionContext
-			? { projectInstructionContext: params.projectInstructionContext }
-			: {}),
-		...(providerContextWindow !== undefined ? { providerContextWindow } : {}),
-		resolveModelContextWindow: async (model) => {
-			if (!modelContextWindows.has(model)) {
-				modelContextWindows.set(
-					model,
-					await resolveProviderContextWindow(
-						resilientProvider,
+		const iterationOrchestrator = new IterationOrchestrator({
+			captureSessionEvidence,
+			provider: resilientProvider,
+			providerCapabilities: capabilities,
+			strictCapabilities: params.strictCapabilities === true,
+			servingMember: () => serving.current,
+			turnConfig,
+			...(params.stopWhen ? { stopWhen: params.stopWhen } : {}),
+			...(params.prepareStep ? { prepareStep: params.prepareStep } : {}),
+			...(params.beforeStep ? { beforeStep: params.beforeStep } : {}),
+			...(params.onStepFinish ? { onStepFinish: params.onStepFinish } : {}),
+			...(params.reviewAnswer ? { reviewAnswer: params.reviewAnswer } : {}),
+			...(params.maxAnswerReviews !== undefined
+				? { maxAnswerReviews: params.maxAnswerReviews }
+				: {}),
+			...(params.structuredOutput ? { structuredOutput: params.structuredOutput } : {}),
+			...(params.parkRecordDelayMs !== undefined
+				? { parkRecordDelayMs: params.parkRecordDelayMs }
+				: {}),
+			tools: params.tools,
+			allowedTools: effectiveAllowedTools,
+			recorder: ctx.recorder,
+			toolExecutor,
+			guard,
+			activityStore: ctx.activityStore,
+			emitEvent: eventTranslator.emitEvent,
+			drainPending: () => eventTranslator.drainPending(),
+			abortController: ctx.abortController,
+			log: ctx.log,
+			// Read through the box on every call, so a swap lands on the next
+			// question rather than the next turn.
+			resumeHandler: (request) => approvalPolicy.current.handler(request),
+			takeApprovalPolicyChange: () => approvalPolicy.takeUnannouncedChange(),
+			promptContributions,
+			...(params.steering ? { steering: params.steering } : {}),
+			...(jobNotices ? { jobNotices } : {}),
+			...(awaitedJobs ? { awaitedJobs } : {}),
+			checkpointMgr,
+			planManager: ctx.planManager,
+			taskGateway: taskScheduler,
+			completionInbox: params.completionInbox,
+			taskStore: params.taskStore,
+			// Turn-scoped. An approval is a statement about this turn's work;
+			// carrying one into a later turn would be reuse nobody agreed to.
+			toolGrants: new ToolGrantSet(),
+			// Turn-scoped for the same reason. A repeat count carried into a later
+			// run is a claim about work nobody repeated, and a module-level map
+			// would leak exactly that way.
+			...(params.repeatCallAdvisory === false ? {} : { repeatCalls: new RepeatCallTracker() }),
+			compactionConfig,
+			...(params.inboundMessages ? { inboundMessages: params.inboundMessages } : {}),
+			...(params.resumeFromCheckpoint ? { resumedInput: queuedForThisRun } : {}),
+			...(params.waitForInbound ? { waitForInbound: params.waitForInbound } : {}),
+			...(params.projectInstructionContext
+				? { projectInstructionContext: params.projectInstructionContext }
+				: {}),
+			...(providerContextWindow !== undefined ? { providerContextWindow } : {}),
+			resolveModelContextWindow: async (model) => {
+				if (!modelContextWindows.has(model)) {
+					modelContextWindows.set(
 						model,
-						ctx.abortController.signal,
-						runConfig.timeoutMs,
-						log,
-					),
-				)
-			}
-			return modelContextWindows.get(model)
-		},
-		workingStateManager,
-		taskRouter: params.taskRouter,
-		contextReducer: params.contextReducer,
-		workingMemoryProvider: params.workingMemoryProvider,
-		advisoryCtx,
-		agentBus: params.agentBus,
-		verificationGate,
-		pluginManager: params.pluginManager,
-	})
-
-	const tracer = getTracer()
-
-	// Whether the run reached its settle. Read by the `finally` below, and
-	// the only thing that distinguishes a run that finished from one whose
-	// consumer walked away — see `settleAbandonedRun`.
-	let settled = false
-
-	const runBody = (async function* (): AsyncGenerator<RunEvent, Run> {
-		// Parent explicitly when a caller supplied one. Without this every
-		// run starts its OWN root trace, so a supervisor delegating to three
-		// children produced four disconnected traces instead of one tree —
-		// the same defect that made a 20-turn run show up as 21 roots before
-		// iterations were parented, except across the spawn boundary, where
-		// it is worse: the delegation structure is the thing you most want
-		// to see.
-		const runStartedAt = Date.now()
-
-		// Read before the span is minted, because a parent can only be set
-		// at creation. A resumed run used to start a brand-new trace with no
-		// link to the one that crashed, so the failure and its recovery
-		// could not be put on one timeline — the run id correlated them well
-		// enough to find both by query and not well enough to see a single
-		// waterfall, and for a replay fork (which mints a new run id) not
-		// even that. An explicit caller-supplied parent still wins: it is
-		// the more specific statement about where this run belongs.
-		const resumedTrace = selectedResumeState
-			? selectedResumeState.traceContext
-			: params.resumeFromCheckpoint
-				? await checkpointMgr.readTraceContext(params.resumeFromCheckpoint)
-				: undefined
-
-		const rootSpan = tracer.startSpan(
-			agentRunSpanName(params.agentName),
-			{},
-			parentContext(params.parentSpan ?? resumedTrace),
-		)
-		// Hand the run span to the loop so every iteration parents to it.
-		iterationOrchestrator.setRootSpan(rootSpan)
-		// Every checkpoint from here on records the trace it was taken
-		// inside, so the next resume can join this one.
-		checkpointMgr.setTraceSource(() => serializeSpan(rootSpan))
-		// And every park it records carries an absolute deadline, so an
-		// unanswered approval cannot outlive the worker that asked for it.
-		checkpointMgr.setParkTtl(runConfig.hitlParkTtlMs)
-		// The claim this worker holds, if it took one. Without this hop the
-		// fence exists, the refusal exists, and no checkpoint a RUN writes ever
-		// carries a number — so a stalled worker is refused nowhere.
-		checkpointMgr.setClaimFence(params.claimFence)
-		// A question raised from inside a tool becomes a real checkpoint
-		// here. It used to park under a synthetic id nothing ever wrote, so
-		// the checkpoint did not exist: nothing on disk said a human owed
-		// this run an answer, and a remote host could not observe the
-		// question at all.
-		questionParks.bind({
-			record: async (question) => {
-				try {
-					const checkpoint = await checkpointMgr.create(ctx.runMgr, ctx.runMgr.currentIteration)
-					const parked = await checkpointMgr.park(checkpoint, {
-						type: 'user_question',
-						runId: ctx.runMgr.id,
-						checkpointId: checkpoint.id,
-						question,
-					})
-					await eventTranslator.emitEvent({
-						type: 'user_question_asked',
-						runId: ctx.runMgr.id,
-						checkpointId: parked.id,
-						questionId: question.questionId,
-						question: question.question,
-					})
-					return parked.id
-				} catch (err) {
-					// A store that cannot record the park must not take the
-					// tool down with it: the in-process await is still valid
-					// and only the cross-process handoff is lost. Loudly,
-					// because a host building an approval queue from durable
-					// state will not see this question.
-					ctx.log.error('Failed to record a question park — it is not resumable', {
-						[NAMZU.RUN_ID]: ctx.runMgr.id,
-						'namzu.runtime.question_id': question.questionId,
-						'exception.message': err instanceof Error ? err.message : String(err),
-					})
-					return null
+						await resolveProviderContextWindow(
+							resilientProvider,
+							model,
+							ctx.abortController.signal,
+							turnConfig.timeoutMs,
+							log,
+						),
+					)
 				}
+				return modelContextWindows.get(model)
 			},
-			resolve: async (checkpointId, decision) => {
-				await checkpointMgr.unpark(checkpointId, decision).catch((err: unknown) => {
-					ctx.log.error('Failed to clear a recorded question park', {
-						[NAMZU.RUN_ID]: ctx.runMgr.id,
-						'namzu.checkpoint.id': checkpointId,
-						'exception.message': err instanceof Error ? err.message : String(err),
-					})
-					return null
-				})
-				await eventTranslator.emitEvent({
-					type: 'user_question_answered',
-					runId: ctx.runMgr.id,
-					checkpointId,
-					...(decision.action === 'answer_question' && decision.questionId !== undefined
-						? { questionId: decision.questionId }
-						: {}),
-					answered: decision.action === 'answer_question',
-				})
-			},
-		})
-		rootSpan.setAttributes({
-			[NAMZU.RUN_ID]: ctx.runMgr.id,
-			[GENAI.AGENT_NAME]: params.agentName,
-			[GENAI.AGENT_ID]: params.agentId,
-			[GENAI.REQUEST_MODEL]: runConfig.model,
-			[GENAI.SYSTEM]: params.provider.id,
+			workingStateManager,
+			taskRouter: params.taskRouter,
+			contextReducer: params.contextReducer,
+			workingMemoryProvider: params.workingMemoryProvider,
+			advisoryCtx,
+			agentBus: params.agentBus,
+			verificationGate,
+			pluginManager: params.pluginManager,
 		})
 
-		let sandbox: Sandbox | undefined
-		// Decided during checkpoint restore, executed after the sandbox
-		// exists — the approved tools may well need it.
-		let pendingResume: PendingResumePlan | null = null
-		/**
-		 * The cadence park this resume answered, when the decision is one the
-		 * ordinary continue path carries out. See the restore path below.
-		 */
-		let answeredParkId: CheckpointId | undefined
-		/** Tool results recovered from the transcript; see the restore path. */
-		let recoveredResults: ReadonlyMap<string, { result: string; isError: boolean }> = new Map()
-		let emergencyManager: EmergencySaveManager | undefined
+		const tracer = getTracer()
+
+		// Whether the turn reached its settle. Read by the `finally` below, and
+		// the only thing that distinguishes a turn that finished from one whose
+		// consumer walked away — see `settleAbandonedTurn`.
+		let settled = false
+
+		const runBody = (async function* (): AsyncGenerator<SessionEvent, Turn> {
+			// Parent explicitly when a caller supplied one. Without this every
+			// run starts its OWN root trace, so a supervisor delegating to three
+			// children produced four disconnected traces instead of one tree —
+			// the same defect that made a 20-turn run show up as 21 roots before
+			// iterations were parented, except across the spawn boundary, where
+			// it is worse: the delegation structure is the thing you most want
+			// to see.
+			const runStartedAt = Date.now()
+
+			// Read before the span is minted, because a parent can only be set
+			// at creation. A resumed turn used to start a brand-new trace with no
+			// link to the one that crashed, so the failure and its recovery
+			// could not be put on one timeline — the turn id correlated them well
+			// enough to find both by query and not well enough to see a single
+			// waterfall, and for a replay fork (which mints a new turn id) not
+			// even that. An explicit caller-supplied parent still wins: it is
+			// the more specific statement about where this turn belongs.
+			const resumedTrace = selectedResumeState
+				? selectedResumeState.traceContext
+				: params.resumeFromCheckpoint
+					? await checkpointMgr.readTraceContext(params.resumeFromCheckpoint)
+					: undefined
 
-		try {
-			await ctx.runMgr.init()
-
-			// A consumer coming back gets what it missed BEFORE the run says
-			// anything new, which is the only order that lets it fold one
-			// stream into one state. It has to follow `init()` — that is what
-			// binds the store and reads the log's head — and precede every
-			// emit below.
-			if (params.eventCursor) {
-				yield* catchUpFromCursor(
-					ctx.runMgr,
-					params.eventCursor,
-					params.onEventReplay,
-					params.claimFence,
-					(error) => {
-						ctx.log.warn('Replay observer failed', {
-							'exception.message': toErrorMessage(error),
+			const rootSpan = tracer.startSpan(
+				agentTurnSpanName(params.agentName),
+				{},
+				parentContext(params.parentSpan ?? resumedTrace),
+			)
+			// Hand the turn span to the loop so every iteration parents to it.
+			iterationOrchestrator.setRootSpan(rootSpan)
+			// Every checkpoint from here on records the trace it was taken
+			// inside, so the next resume can join this one.
+			checkpointMgr.setTraceSource(() => serializeSpan(rootSpan))
+			// And every park it records carries an absolute deadline, so an
+			// unanswered approval cannot outlive the worker that asked for it.
+			checkpointMgr.setParkTtl(turnConfig.hitlParkTtlMs)
+			// A question raised from inside a tool becomes a real checkpoint
+			// here. It used to park under a synthetic id nothing ever wrote, so
+			// the checkpoint did not exist: nothing on disk said a human owed
+			// this turn an answer, and a remote host could not observe the
+			// question at all.
+			questionParks.bind({
+				record: async (question) => {
+					try {
+						const checkpoint = await checkpointMgr.create(
+							ctx.recorder,
+							ctx.recorder.currentIteration,
+						)
+						await checkpointMgr.park(checkpoint, {
+							type: 'user_question',
+							sessionId: ctx.sessionId,
+							turnId: ctx.turnId,
+							checkpointId: checkpoint.id,
+							question,
 						})
-					},
-				)
-			}
-
-			ctx.log.info('Starting query', {
-				[NAMZU.RUN_ID]: ctx.runMgr.id,
-				'namzu.runtime.agent': params.agentName,
-				[GENAI.REQUEST_MODEL]: runConfig.model,
-				'namzu.runtime.token_budget': runConfig.tokenBudget,
-				'namzu.runtime.activity_tracking': ctx.activityStore.enabled,
-				'namzu.runtime.permission_mode': ctx.permissionMode.current,
-				'namzu.runtime.resume_from_checkpoint': params.resumeFromCheckpoint ?? null,
-			})
-
-			// The operator's prompt, before the model sees it. A hook may
-			// refuse it — the run ends here, before a prompt is built — or add
-			// to what the model is told; the addition rides the prompt as a
-			// dynamic contribution so every iteration of the run carries it.
-			if (params.pluginManager) {
-				const hookResults = await params.pluginManager.executeHooks(
-					'user_prompt_submit',
-					{
-						runId: ctx.runId,
-						sessionId: params.sessionId,
-						prompt: lastUserPrompt(initialMessages),
-						signal: ctx.abortController.signal,
-					},
-					eventTranslator.emitEvent,
-				)
-				const annotations = applyLifecycleHookResults('user_prompt_submit', hookResults)
-				if (annotations.length > 0) {
-					promptContributions.replace({
-						id: 'hooks:user_prompt_submit',
-						placement: 'dynamic',
-						render: () => `Context from the operator's hooks:\n\n${annotations.join('\n\n')}`,
+						await eventTranslator.emitEvent({
+							type: 'user_question_asked',
+							checkpointId: checkpoint.id,
+							questionId: question.questionId,
+							question: question.question,
+						})
+						return checkpoint.id
+					} catch (err) {
+						// A store that cannot record the park must not take the
+						// tool down with it: the in-process await is still valid
+						// and only the cross-process handoff is lost. Loudly,
+						// because a host building an approval queue from durable
+						// state will not see this question.
+						ctx.log.error('Failed to record a question park — it is not resumable', {
+							[NAMZU.TURN_ID]: ctx.recorder.turnId,
+							'namzu.runtime.question_id': question.questionId,
+							'exception.message': err instanceof Error ? err.message : String(err),
+						})
+						return null
+					}
+				},
+				resolve: async (checkpointId, decision) => {
+					await checkpointMgr.unpark(checkpointId, decision).catch((err: unknown) => {
+						ctx.log.error('Failed to clear a recorded question park', {
+							[NAMZU.TURN_ID]: ctx.recorder.turnId,
+							'namzu.checkpoint.id': checkpointId,
+							'exception.message': err instanceof Error ? err.message : String(err),
+						})
+						return null
 					})
-				}
-				yield* eventTranslator.drainPending()
-			}
-
-			const contextLevel = params.contextLevel ?? 'full'
-			const cacheInput = {
-				systemPrompt: params.systemPrompt,
-				persona: params.persona,
-				skills: params.skills,
-				basePrompt: contextLevel === 'full' ? params.basePrompt : undefined,
-				tools: params.tools,
-				allowedTools: effectiveAllowedTools,
-				runtimeContext: params.runtimeContext,
-				contributions: promptContributions,
-			}
-
-			const segments: PromptSegments = promptCache
-				? promptCache.getSystemPromptSegmented(cacheInput, contextLevel, params.workingDirectory)
-				: promptBuilder.buildSegmented(contextLevel, params.workingDirectory)
-
-			ctx.log.info('Prompt segments assembled', {
-				'namzu.runtime.static_length': segments.static.length,
-				'namzu.runtime.dynamic_length': segments.dynamic.length,
+					await eventTranslator.emitEvent({
+						type: 'user_question_answered',
+						checkpointId,
+						...(decision.action === 'answer_question' && decision.questionId !== undefined
+							? { questionId: decision.questionId }
+							: {}),
+						answered: decision.action === 'answer_question',
+					})
+				},
+			})
+			rootSpan.setAttributes({
+				[GENAI.CONVERSATION_ID]: ctx.sessionId,
+				[NAMZU.TURN_ID]: ctx.recorder.turnId,
+				...(params.parentSessionId !== undefined && {
+					[NAMZU.SESSION_PARENT_ID]: params.parentSessionId,
+				}),
+				[GENAI.AGENT_NAME]: params.agentName,
+				[GENAI.AGENT_ID]: params.agentId,
+				[GENAI.REQUEST_MODEL]: turnConfig.model,
+				[GENAI.SYSTEM]: params.provider.id,
 			})
 
-			const pushSystemMessages = (): void => {
-				ctx.runMgr.pushMessage(createSystemMessage(segments.static, 'cache'))
-				if (segments.dynamic.length > 0) {
-					ctx.runMgr.pushMessage(createSystemMessage(segments.dynamic, 'ephemeral'))
+			let sandbox: Sandbox | undefined
+			// Decided during checkpoint restore, executed after the sandbox
+			// exists — the approved tools may well need it.
+			let pendingResume: PendingResumePlan | null = null
+			/** The record id of the parked assistant message the plan re-appends. */
+			let pendingResumeAssistantId: MessageId | undefined
+			/**
+			 * The cadence park this resume answered, when the decision is one the
+			 * ordinary continue path carries out. See the restore path below.
+			 */
+			let answeredParkId: CheckpointId | undefined
+			/** Tool results recovered from the transcript; see the restore path. */
+			let recoveredResults: ReadonlyMap<string, { result: string; isError: boolean }> = new Map()
+
+			try {
+				// A consumer coming back gets what it missed BEFORE the turn says
+				// anything new, which is the only order that lets it fold one
+				// stream into one state. It follows the prelude's `open()` — which
+				// took the lease and read the log's head — and precedes every emit
+				// below.
+				if (params.eventCursor) {
+					yield* catchUpFromCursor(
+						ctx.recorder,
+						params.eventCursor,
+						params.onEventReplay,
+						(error) => {
+							ctx.log.warn('Replay observer failed', {
+								'exception.message': toErrorMessage(error),
+							})
+						},
+					)
 				}
-			}
 
-			if (params.resumeFromCheckpoint) {
-				const checkpoint = await checkpointMgr.restore(params.resumeFromCheckpoint)
-				const projectedCheckpoint = {
-					...checkpoint,
-					messages: projectStateBearingHistory(checkpoint.messages, {
-						pinCompaction: false,
-					}),
-				}
-				await eventTranslator.emitEvent({
-					type: 'run_resuming',
-					runId: ctx.runMgr.id,
-					fromCheckpointId: checkpoint.id,
+				ctx.log.info('Starting query', {
+					[NAMZU.TURN_ID]: ctx.recorder.turnId,
+					'namzu.runtime.agent': params.agentName,
+					[GENAI.REQUEST_MODEL]: turnConfig.model,
+					'namzu.runtime.token_budget': turnConfig.tokenBudget,
+					'namzu.runtime.activity_tracking': ctx.activityStore.enabled,
+					'namzu.runtime.permission_mode': ctx.permissionMode.current,
+					'namzu.runtime.resume_from_checkpoint': params.resumeFromCheckpoint ?? null,
 				})
-				yield* eventTranslator.drainPending()
 
-				// Budgets are properties of the RUN, not of the process hosting
-				// it. The checkpoint already carried all three; they were
-				// written and then discarded on the way back in, so a run
-				// recalled at $4.80 of a $5 cap came back with a fresh $5 and
-				// a fresh timeout clock. Restore before the first iteration so
-				// a resumed run that is already over budget stops immediately.
-				ctx.runMgr.restoreUsage(
-					checkpoint.tokenUsage,
-					checkpoint.costInfo,
-					checkpoint.guardState.iterationCount,
-				)
-				guard.restoreElapsed(checkpoint.guardState.elapsedMs)
-
-				// Adopt the working state the earlier summary was built from.
-				// The `[COMPACTED CONTEXT]` block below is preserved precisely
-				// because it is the only surviving record of the history the
-				// first pass deleted — and without this, the NEXT compaction
-				// would drop it and replace it with a summary covering only
-				// what happened after the resume, silently losing the run's
-				// first hour.
-				if (workingStateManager && checkpoint.workingState && compactionConfig) {
-					const revived = restoreWorkingState(checkpoint.workingState, compactionConfig)
-					workingStateManager.replaceState(revived.getState())
-					ctx.log.info('Restored compaction working state from checkpoint', {
-						[NAMZU.RUN_ID]: ctx.runMgr.id,
-						'namzu.checkpoint.id': checkpoint.id,
-						'namzu.runtime.slots': workingStateManager.slotCount(),
-					})
+				// The operator's prompt, before the model sees it. A hook may
+				// refuse it — the turn ends here, before a prompt is built — or add
+				// to what the model is told; the addition rides the prompt as a
+				// dynamic contribution so every iteration of the turn carries it.
+				if (params.pluginManager) {
+					const hookResults = await params.pluginManager.executeHooks(
+						'user_prompt_submit',
+						{
+							turnId: ctx.turnId,
+							sessionId: params.sessionId,
+							prompt: lastUserPrompt(initialMessages),
+							signal: ctx.abortController.signal,
+						},
+						eventTranslator.emitEvent,
+					)
+					const annotations = applyLifecycleHookResults('user_prompt_submit', hookResults)
+					if (annotations.length > 0) {
+						promptContributions.replace({
+							id: 'hooks:user_prompt_submit',
+							placement: 'dynamic',
+							render: () => `Context from the operator's hooks:\n\n${annotations.join('\n\n')}`,
+						})
+					}
+					yield* eventTranslator.drainPending()
 				}
-				ctx.log.info('Restored budgets from checkpoint', {
-					[NAMZU.RUN_ID]: ctx.runMgr.id,
-					'namzu.checkpoint.id': checkpoint.id,
-					'namzu.usage.total_tokens': checkpoint.tokenUsage.totalTokens,
-					'namzu.runtime.total_cost': checkpoint.costInfo.totalCost,
-					[NAMZU.ITERATION]: checkpoint.guardState.iterationCount,
-					'namzu.runtime.elapsed_ms': checkpoint.guardState.elapsedMs,
-				})
-
-				pushSystemMessages()
-
-				// A human answered the park in a different process; apply that
-				// answer to the tool calls they were actually shown. Without
-				// this the repair below throws the approval away and the model
-				// re-decides, so "yes, delete that row" degrades into "ask
-				// again and hope it asks for the same thing".
-				pendingResume =
-					params.pendingDecision && projectedCheckpoint.pending
-						? planPendingResume(projectedCheckpoint, params.pendingDecision, ctx.log)
-						: null
 
-				// The park this resume ANSWERS even though there is no plan to
-				// carry the decision out through.
-				//
-				// `planPendingResume` covers the two arms whose decision has to
-				// reach something — the calls a `tool_review` park is about, the
-				// tool a `user_question` park is inside. An `iteration_checkpoint`
-				// park has neither, so it returns no plan, and the unpark further
-				// down — which ran only when there was one — never fired for it.
-				// A run that parked on the cadence, was resumed with
-				// `{action: 'continue'}` and went on to finish its work therefore
-				// kept reporting an OUTSTANDING park to `findPendingCheckpoint`,
-				// so a second resume of the finished run was refused with
-				// `awaiting-decision` for a decision already taken, and because
-				// `prune` skips an unresolved park the row could no longer be
-				// collected by anything.
-				//
-				// The decision IS carried out here — continuing is exactly what
-				// the loop below does, and a plan verdict is the answer to the
-				// question the plan park asked — so the park is resolved at the
-				// same point and with the same meaning "resolved" carries
-				// everywhere else: the record stays, and only its pending state
-				// ends. A `pause` is deliberately not resolved: it holds the
-				// park rather than answering it, which is how the live path
-				// treats it too. `answersParkOf` is the whole map, park type to
-				// answering decision, so an arm cannot go missing by being
-				// absent from a condition again — which is how the plan arm
-				// leaked a finished run's park.
-				//
-				// Resolving it does not depend on the resumed process being able
-				// to act on it, and that is deliberate: the plan's own fate is a
-				// separate defect (nothing restores a plan on the resume path at
-				// all, so the new process has none to approve, execute or
-				// reject) and making the resolution wait for it would leave the
-				// row outstanding for exactly the runs that need it cleared.
-				const parked = projectedCheckpoint.pending
-				answeredParkId =
-					params.pendingDecision &&
-					parked !== undefined &&
-					parked.resolvedAt === undefined &&
-					answersParkOf(parked.request.type, params.pendingDecision)
-						? projectedCheckpoint.id
-						: undefined
+				const contextLevel = params.contextLevel ?? 'full'
+				const cacheInput = {
+					systemPrompt: params.systemPrompt,
+					persona: params.persona,
+					skills: params.skills,
+					basePrompt: contextLevel === 'full' ? params.basePrompt : undefined,
+					tools: params.tools,
+					allowedTools: effectiveAllowedTools,
+					runtimeContext: params.runtimeContext,
+					contributions: promptContributions,
+				}
 
-				// Recover completed observations and explicitly unknown outcomes.
-				// A recorded start is not proof that its external effect failed.
-				const unanswered = interruptedToolCalls(projectedCheckpoint.messages)
-				recoveredResults =
-					unanswered.length > 0
-						? await recoverCompletedCalls(ctx.runMgr, unanswered, ctx.log, {
-								answers: pendingResume?.answers,
-								signal: ctx.abortController.signal,
-							})
-						: new Map()
+				const segments: PromptSegments = promptCache
+					? promptCache.getSystemPromptSegmented(cacheInput, contextLevel, params.workingDirectory)
+					: promptBuilder.buildSegmented(contextLevel, params.workingDirectory)
 
-				// Preserve both known and unknown outcomes before continuing calls
-				// whose absence of a start was established by a complete scan.
-				if (!pendingResume && recoveredResults.size > 0) {
-					pendingResume = planCrashResume(projectedCheckpoint, recoveredResults, ctx.log)
-				}
+				ctx.log.info('Prompt segments assembled', {
+					'namzu.runtime.static_length': segments.static.length,
+					'namzu.runtime.dynamic_length': segments.dynamic.length,
+				})
 
-				// An incomplete turn with a durable owner is NOT abandoned. A
-				// pending decision or crash-resume plan re-appends that exact
-				// assistant with real/denied/recovered results below. Remove it
-				// from the generic pass so no synthetic result competes with the
-				// authority that still owns the call. Everything else is abandoned
-				// history and is repaired conservatively rather than deleted.
-				const abandonedCheckpointMessages = pendingResume
-					? withoutOwnedResumeTurn(projectedCheckpoint.messages, pendingResume.assistant)
-					: projectedCheckpoint.messages
-				const checkpointRepair = repairToolMessageHistory(abandonedCheckpointMessages)
-				const restoredMessages = checkpointRepair.messages
-				if (toolHistoryRepairChanged(checkpointRepair.report)) {
-					pendingHistoryRepairs.push({
-						source: 'abandoned-checkpoint',
-						report: checkpointRepair.report,
-					})
-					ctx.log.warn('Repaired abandoned tool history while restoring a checkpoint', {
-						[NAMZU.RUN_ID]: ctx.runMgr.id,
-						'namzu.checkpoint.id': checkpoint.id,
-						'namzu.history.source': 'abandoned-checkpoint',
-						'namzu.history.duplicate_tool_results_removed':
-							checkpointRepair.report.duplicateToolResultsRemoved,
-						'namzu.history.orphaned_tool_results_removed':
-							checkpointRepair.report.orphanedToolResultsRemoved,
-						'namzu.history.synthetic_tool_results_inserted':
-							checkpointRepair.report.syntheticToolResultsInserted,
+				// The prompt floor is rebuilt every turn and never recorded:
+				// `turn_started.systemPrompt` carries it.
+				const pushSystemMessages = (): void => {
+					ctx.recorder.pushMessage(createSystemMessage(segments.static, 'cache'), {
+						transient: true,
 					})
+					if (segments.dynamic.length > 0) {
+						ctx.recorder.pushMessage(createSystemMessage(segments.dynamic, 'ephemeral'), {
+							transient: true,
+						})
+					}
+				}
+				const pushRestored = (message: Message, ids: ReadonlyMap<Message, MessageId>): void => {
+					const messageId = ids.get(message)
+					ctx.recorder.pushMessage(message, messageId ? { messageId } : {})
 				}
 
-				// The ledger is process state and a resumed run starts with an empty
-				// one, so until something reads a file again this run knows nothing
-				// about files the conversation already wrote in full — and re-reads
-				// them. Rebuilt from the REPAIRED history, which is what the model
-				// is about to be shown, rather than from the checkpoint's own
-				// messages — plus whatever of the owned resume turn actually ran,
-				// which the repaired list does not carry; see
-				// `withOwnedResumeOutcomes`. Reads no file's content: every
-				// fingerprint recovered here is still checked against the real one
-				// at mutation time.
-				try {
-					await toolExecutor.seedFileObservations(
-						pendingResume
-							? withOwnedResumeOutcomes(restoredMessages, pendingResume.assistant, recoveredResults)
-							: restoredMessages,
-						params.sandboxProvider !== undefined,
+				if (params.resumeFromCheckpoint) {
+					const checkpoint = await checkpointMgr.restore(params.resumeFromCheckpoint)
+					const projectedCheckpoint = {
+						...checkpoint,
+						messages: projectStateBearingHistory(checkpoint.messages, {
+							pinCompaction: false,
+						}),
+					}
+					await eventTranslator.resumeTurn(checkpoint.id)
+					yield* eventTranslator.drainPending()
+
+					// Budgets are properties of the TURN, not of the process hosting
+					// it. The checkpoint already carried all three; they were
+					// written and then discarded on the way back in, so a turn
+					// recalled at $4.80 of a $5 cap came back with a fresh $5 and
+					// a fresh timeout clock. Restore before the first iteration so
+					// a resumed turn that is already over budget stops immediately.
+					ctx.recorder.restoreUsage(
+						checkpoint.document.tokenUsage,
+						checkpoint.document.costInfo,
+						checkpoint.document.guards.iteration,
 					)
-				} catch (err: unknown) {
-					// A ledger that could not be rebuilt is the empty one every resume
-					// used to get, so the run continues without its witnesses and the
-					// model reads what it needs. Failing the resume over it would trade
-					// a conversation that works for one that does not, to protect an
-					// optimisation. Said out loud all the same, because a seeding that
-					// failed and one that found nothing are otherwise the same silence.
-					ctx.log.debug('Could not rebuild the file observation ledger on resume', {
-						[NAMZU.RUN_ID]: ctx.runMgr.id,
+					guard.restoreElapsed(checkpoint.document.guards.elapsedMs)
+
+					// Adopt the working state the earlier summary was built from.
+					// The `[COMPACTED CONTEXT]` block below is preserved precisely
+					// because it is the only surviving record of the history the
+					// first pass deleted — and without this, the NEXT compaction
+					// would drop it and replace it with a summary covering only
+					// what happened after the resume, silently losing the turn's
+					// first hour.
+					if (workingStateManager && checkpoint.document.workingState && compactionConfig) {
+						const revived = restoreWorkingState(checkpoint.document.workingState, compactionConfig)
+						workingStateManager.replaceState(revived.getState())
+						ctx.log.info('Restored compaction working state from checkpoint', {
+							[NAMZU.TURN_ID]: ctx.recorder.turnId,
+							'namzu.checkpoint.id': checkpoint.id,
+							'namzu.runtime.slots': workingStateManager.slotCount(),
+						})
+					}
+					ctx.log.info('Restored budgets from checkpoint', {
+						[NAMZU.TURN_ID]: ctx.recorder.turnId,
 						'namzu.checkpoint.id': checkpoint.id,
-						'exception.message': err instanceof Error ? err.message : String(err),
+						'namzu.usage.total_tokens': checkpoint.document.tokenUsage.totalTokens,
+						'namzu.runtime.total_cost': checkpoint.document.costInfo.totalCost,
+						[NAMZU.ITERATION]: checkpoint.document.guards.iteration,
+						'namzu.runtime.elapsed_ms': checkpoint.document.guards.elapsedMs,
 					})
-				}
 
-				for (const msg of restoredMessages) {
-					if (msg.role === 'system') {
-						// Re-push the FRESH static/dynamic floor (done above) but PRESERVE
-						// the two system messages that carry irreplaceable run state: the
-						// `[COMPACTED CONTEXT]` summary is the only surviving record of the
-						// older history a compaction pass deleted, and the working-memory
-						// slot pins the produced-artifact ledger. Dropping every system
-						// message on restore silently lost both on resume.
-						if (isCompactionMessage(msg.content) || isWorkingMemoryMessage(msg.content)) {
-							ctx.runMgr.pushMessage(msg)
+					pushSystemMessages()
+
+					// A human answered the park in a different process; apply that
+					// answer to the tool calls they were actually shown. Without
+					// this the repair below throws the approval away and the model
+					// re-decides, so "yes, delete that row" degrades into "ask
+					// again and hope it asks for the same thing".
+					pendingResume =
+						params.pendingDecision && projectedCheckpoint.pending
+							? planPendingResume(projectedCheckpoint, params.pendingDecision, ctx.log)
+							: null
+
+					// The park this resume ANSWERS even though there is no plan to
+					// carry the decision out through.
+					//
+					// `planPendingResume` covers the two arms whose decision has to
+					// reach something — the calls a `tool_review` park is about, the
+					// tool a `user_question` park is inside. An `iteration_checkpoint`
+					// park has neither, so it returns no plan, and the unpark further
+					// down — which ran only when there was one — never fired for it.
+					// A turn that parked on the cadence, was resumed with
+					// `{action: 'continue'}` and went on to finish its work therefore
+					// kept reporting an OUTSTANDING park to `findPendingCheckpoint`,
+					// so a second resume of the finished turn was refused with
+					// `awaiting-decision` for a decision already taken, and because
+					// `prune` skips an unresolved park the row could no longer be
+					// collected by anything.
+					//
+					// The decision IS carried out here — continuing is exactly what
+					// the loop below does, and a plan verdict is the answer to the
+					// question the plan park asked — so the park is resolved at the
+					// same point and with the same meaning "resolved" carries
+					// everywhere else: the record stays, and only its pending state
+					// ends. A `pause` is deliberately not resolved: it holds the
+					// park rather than answering it, which is how the live path
+					// treats it too. `answersParkOf` is the whole map, park type to
+					// answering decision, so an arm cannot go missing by being
+					// absent from a condition again — which is how the plan arm
+					// leaked a finished turn's park.
+					//
+					// Resolving it does not depend on the resumed process being able
+					// to act on it, and that is deliberate: the plan's own fate is a
+					// separate defect (nothing restores a plan on the resume path at
+					// all, so the new process has none to approve, execute or
+					// reject) and making the resolution wait for it would leave the
+					// row outstanding for exactly the turns that need it cleared.
+					const parked = projectedCheckpoint.pending
+					answeredParkId =
+						params.pendingDecision &&
+						parked !== undefined &&
+						parked.resolvedAt === undefined &&
+						answersParkOf(parked.request.type, params.pendingDecision)
+							? projectedCheckpoint.id
+							: undefined
+
+					// Recover completed observations and explicitly unknown outcomes.
+					// A recorded start is not proof that its external effect failed.
+					const unanswered = interruptedToolCalls(projectedCheckpoint.messages)
+					recoveredResults =
+						unanswered.length > 0
+							? await recoverCompletedCalls(ctx.recorder, unanswered, ctx.log, {
+									answers: pendingResume?.answers,
+									signal: ctx.abortController.signal,
+								})
+							: new Map()
+
+					// Preserve both known and unknown outcomes before continuing calls
+					// whose absence of a start was established by a complete scan.
+					if (!pendingResume && recoveredResults.size > 0) {
+						pendingResume = planCrashResume(projectedCheckpoint, recoveredResults, ctx.log)
+					}
+
+					// An incomplete turn with a durable owner is NOT abandoned. A
+					// pending decision or crash-resume plan re-appends that exact
+					// assistant with real/denied/recovered results below. Remove it
+					// from the generic pass so no synthetic result competes with the
+					// authority that still owns the call. Everything else is abandoned
+					// history and is repaired conservatively rather than deleted.
+					pendingResumeAssistantId = pendingResume
+						? checkpoint.messageIds.get(pendingResume.assistant)
+						: undefined
+					const abandonedCheckpointMessages = pendingResume
+						? withoutOwnedResumeTurn(projectedCheckpoint.messages, pendingResume.assistant)
+						: projectedCheckpoint.messages
+					const checkpointRepair = repairToolMessageHistory(abandonedCheckpointMessages)
+					const restoredMessages = checkpointRepair.messages
+					if (toolHistoryRepairChanged(checkpointRepair.report)) {
+						pendingHistoryRepairs.push({
+							source: 'abandoned-checkpoint',
+							report: checkpointRepair.report,
+						})
+						ctx.log.warn('Repaired abandoned tool history while restoring a checkpoint', {
+							[NAMZU.TURN_ID]: ctx.recorder.turnId,
+							'namzu.checkpoint.id': checkpoint.id,
+							'namzu.history.source': 'abandoned-checkpoint',
+							'namzu.history.duplicate_tool_results_removed':
+								checkpointRepair.report.duplicateToolResultsRemoved,
+							'namzu.history.orphaned_tool_results_removed':
+								checkpointRepair.report.orphanedToolResultsRemoved,
+							'namzu.history.synthetic_tool_results_inserted':
+								checkpointRepair.report.syntheticToolResultsInserted,
+						})
+					}
+
+					// The ledger is process state and a resumed turn starts with an empty
+					// one, so until something reads a file again this turn knows nothing
+					// about files the conversation already wrote in full — and re-reads
+					// them. Rebuilt from the REPAIRED history, which is what the model
+					// is about to be shown, rather than from the checkpoint's own
+					// messages — plus whatever of the owned resume turn actually ran,
+					// which the repaired list does not carry; see
+					// `withOwnedResumeOutcomes`. Reads no file's content: every
+					// fingerprint recovered here is still checked against the real one
+					// at mutation time.
+					try {
+						await toolExecutor.seedFileObservations(
+							pendingResume
+								? withOwnedResumeOutcomes(
+										restoredMessages,
+										pendingResume.assistant,
+										recoveredResults,
+									)
+								: restoredMessages,
+							params.sandboxProvider !== undefined,
+						)
+					} catch (err: unknown) {
+						// A ledger that could not be rebuilt is the empty one every resume
+						// used to get, so the turn continues without its witnesses and the
+						// model reads what it needs. Failing the resume over it would trade
+						// a conversation that works for one that does not, to protect an
+						// optimisation. Said out loud all the same, because a seeding that
+						// failed and one that found nothing are otherwise the same silence.
+						ctx.log.debug('Could not rebuild the file observation ledger on resume', {
+							[NAMZU.TURN_ID]: ctx.recorder.turnId,
+							'namzu.checkpoint.id': checkpoint.id,
+							'exception.message': err instanceof Error ? err.message : String(err),
+						})
+					}
+
+					for (const msg of restoredMessages) {
+						if (msg.role === 'system') {
+							// Re-push the FRESH static/dynamic floor (done above) but PRESERVE
+							// the two system messages that carry irreplaceable run state: the
+							// `[COMPACTED CONTEXT]` summary is the only surviving record of the
+							// older history a compaction pass deleted, and the working-memory
+							// slot pins the produced-artifact ledger. Dropping every system
+							// message on restore silently lost both on resume.
+							if (isCompactionMessage(msg.content) || isWorkingMemoryMessage(msg.content)) {
+								pushRestored(msg, checkpoint.messageIds)
+							}
+							continue
 						}
-						continue
+						pushRestored(msg, checkpoint.messageIds)
 					}
-					ctx.runMgr.pushMessage(msg)
-				}
 
-				// The queue, on the resume path too. It is drained
-				// unconditionally above, so leaving this out would take a
-				// host's "start with this" off the record and deliver it
-				// nowhere — the one outcome a durable queue must not have.
-				//
-				// AFTER the restored history rather than before it: on a
-				// resume the conversation already exists, and a message left
-				// for "the next run" is the newest thing said, not the oldest.
-				for (const queued of queuedForThisRun) ctx.runMgr.pushMessage(queued)
-			} else if (params.continuationMode) {
-				for (const msg of initialMessages) {
-					ctx.runMgr.pushMessage(msg)
-				}
-			} else {
-				pushSystemMessages()
-				for (const msg of initialMessages) {
-					if (msg.role === 'system') {
-						// A fresh run rebuilds its current static/dynamic prompt above,
-						// so arbitrary historical system messages stay out. These two
-						// are different: they are conversation STATE, and dropping them
-						// deletes the only surviving record of compacted history or the
-						// produced-artifact ledger. A compaction summary arriving from a
-						// prior run is pinned because this new WorkingStateManager cannot
-						// prove it has reconstructed equivalent state yet.
-						if (isCompactionMessage(msg.content)) {
-							ctx.runMgr.pushMessage({ ...msg, retain: true })
-						} else if (isWorkingMemoryMessage(msg.content)) {
-							ctx.runMgr.pushMessage(msg)
+					// The queue, on the resume path too. It is drained
+					// unconditionally above, so leaving this out would take a
+					// host's "start with this" off the record and deliver it
+					// nowhere — the one outcome a durable queue must not have.
+					//
+					// AFTER the restored history rather than before it: on a
+					// resume the conversation already exists, and a message left
+					// for "the next turn" is the newest thing said, not the oldest.
+					for (const queued of queuedForThisRun) ctx.recorder.pushMessage(queued)
+				} else if (params.continuationMode) {
+					for (const msg of initialMessages) pushRestored(msg, historyIds)
+				} else {
+					pushSystemMessages()
+					for (const msg of initialMessages) {
+						if (msg.role === 'system') {
+							// A fresh turn rebuilds its current static/dynamic prompt above,
+							// so arbitrary historical system messages stay out. These two
+							// are different: they are conversation STATE, and dropping them
+							// deletes the only surviving record of compacted history or the
+							// produced-artifact ledger. A compaction summary arriving from a
+							// prior turn is pinned because this new WorkingStateManager cannot
+							// prove it has reconstructed equivalent state yet.
+							if (isCompactionMessage(msg.content)) {
+								// Pinned by the prelude already; a copy would lose its record id.
+								pushRestored(msg.retain ? msg : { ...msg, retain: true }, historyIds)
+							} else if (isWorkingMemoryMessage(msg.content)) {
+								pushRestored(msg, historyIds)
+							}
+							continue
 						}
-						continue
+						pushRestored(msg, historyIds)
 					}
-					ctx.runMgr.pushMessage(msg)
 				}
-			}
-			// Fresh and continuation histories seed the same operator state.
-			// A worker report or project-policy message may have role `user`,
-			// but it must not become the task or an operator requirement.
-			// Resume restores this state and adopts only its newer queue arrivals.
-			if (!params.resumeFromCheckpoint && workingStateManager) {
-				let isFirstUserMessage = true
-				for (const message of initialMessages) {
-					if (!isOperatorUserMessage(message) || !message.content.trim()) continue
-					extractFromUserMessage(workingStateManager, message.content, isFirstUserMessage)
-					isFirstUserMessage = false
+				// Fresh and continuation histories seed the same operator state.
+				// A worker report or project-policy message may have role `user`,
+				// but it must not become the task or an operator requirement.
+				// Resume restores this state and adopts only its newer queue arrivals.
+				if (!params.resumeFromCheckpoint && workingStateManager) {
+					let isFirstUserMessage = true
+					for (const message of initialMessages) {
+						if (!isOperatorUserMessage(message) || !message.content.trim()) continue
+						extractFromUserMessage(workingStateManager, message.content, isFirstUserMessage)
+						isFirstUserMessage = false
+					}
 				}
-			}
-
-			const assembledPrompt =
-				segments.dynamic.length > 0
-					? `${segments.static}\n\n---\n\n${segments.dynamic}`
-					: segments.static
-
-			ctx.runMgr.markRunning()
-			await eventTranslator.emitEvent({
-				type: 'run_started',
-				runId: ctx.runMgr.id,
-				systemPrompt: assembledPrompt,
-			})
-			yield* eventTranslator.drainPending()
-
-			// Pre-run materialization can observe cancellation before RunContext
-			// exists. The exact input has now been seeded and the run is writable;
-			// hand the cancellation to the normal terminal path before invoking
-			// any host callback, guardrail, plugin, sandbox, or provider. Those
-			// boundaries are not all cooperative and must not regain withdrawn
-			// authority merely because the run record still had to be created.
-			ctx.abortController.signal.throwIfAborted()
-
-			// Handed over here, and the position is load-bearing in three
-			// directions. It has to follow `wirePlanManager`, or a host that
-			// builds its plan in this callback does it into silence. It has to
-			// follow `runMgr.init()` and `run_started`, because plan events append
-			// to that durable run. It also has to follow the pre-model abort fence:
-			// a callback invoked after attachment resolution observed cancellation
-			// would regain withdrawn authority and could replace the cancellation
-			// with its own failure. This is still before the iteration loop, which
-			// is the guarantee the callback makes.
-			params.onContextCreated?.({ planManager: ctx.planManager })
-
-			// The box is handed out HERE, after `run_started`, and the position
-			// is load-bearing rather than tidy. It moved twice:
-			//
-			//  1. Beside the box's construction — a host that called `set`
-			//     synchronously reached `eventTranslator` inside its temporal
-			//     dead zone and killed the run before it started.
-			//  2. Beside the translator's construction — the translator existed,
-			//     but the run directory did not, so the durable append hit
-			//     ENOENT on `transcript.jsonl`.
-			//
-			// Both were found by the test that takes the box and immediately
-			// swaps the policy, which is not an exotic host: it is the shape of
-			// "start unattended" wiring. A policy change is durably recorded
-			// before it takes effect, so the handout cannot precede the run
-			// being writable.
-			params.onApprovalPolicy?.(approvalPolicy)
-
-			// History repair happens before the run manager sees the first model
-			// request, but its durable event cannot precede run_started: there is no
-			// writable run log until that event initializes it. Emit the measured
-			// counts here, still before any provider call, so hosts can tell that the
-			// model received a repaired projection rather than the raw history.
-			for (const repair of pendingHistoryRepairs) {
-				await eventTranslator.emitEvent({
-					type: 'message_history_repaired',
-					runId: ctx.runMgr.id,
-					source: repair.source,
-					...repair.report,
-				})
-				yield* eventTranslator.drainPending()
-			}
 
-			// Surface capability degradation to the host as run events —
-			// explicit, not silent (the log.warn above fires at setup time;
-			// this is the machine-readable channel).
-			if (stripToolSurfaces) {
-				await eventTranslator.emitEvent({
-					type: 'capability_warning',
-					runId: ctx.runMgr.id,
-					capability: 'tools',
-					providerId: params.provider.id,
-					message: `Provider '${params.provider.id}' does not support tools — ${registeredToolCount} registered tool(s) were stripped from the prompt and request.`,
-				})
-				yield* eventTranslator.drainPending()
-			}
-			if (attachmentMessageCount > 0) {
-				await eventTranslator.emitEvent({
-					type: 'capability_warning',
-					runId: ctx.runMgr.id,
-					capability: 'vision',
-					providerId: params.provider.id,
-					message: `Provider '${params.provider.id}' does not support vision — image attachments on ${attachmentMessageCount} user message(s) will not reach the model.`,
-				})
-				yield* eventTranslator.drainPending()
-			}
-			if (documentMessageCount > 0) {
-				await eventTranslator.emitEvent({
-					type: 'capability_warning',
-					runId: ctx.runMgr.id,
-					capability: 'documents',
-					providerId: params.provider.id,
-					message: `Provider '${params.provider.id}' does not support documents — document attachments on ${documentMessageCount} user message(s) will not reach the model.`,
-				})
+				const assembledPrompt =
+					segments.dynamic.length > 0
+						? `${segments.static}\n\n---\n\n${segments.dynamic}`
+						: segments.static
+
+				ctx.recorder.markRunning()
+				// A resumed turn already began; `turn_resuming` above continued it.
+				if (!params.resumeFromCheckpoint) {
+					await eventTranslator.beginTurn({
+						systemPrompt: assembledPrompt,
+						...(params.origin ? { origin: params.origin } : {}),
+						...(params.abandonInterrupted ? { abandonInterrupted: true } : {}),
+					})
+				}
 				yield* eventTranslator.drainPending()
-			}
 
-			if (params.pluginManager) {
-				const hookResults = await params.pluginManager.executeHooks(
-					'run_start',
-					{ runId: ctx.runId, signal: ctx.abortController.signal },
-					eventTranslator.emitEvent,
-				)
-				applyLifecycleHookResults('run_start', hookResults)
-				yield* eventTranslator.drainPending()
-			}
+				// Pre-run materialization can observe cancellation before TurnContext
+				// exists. The exact input has now been seeded and the turn is writable;
+				// hand the cancellation to the normal terminal path before invoking
+				// any host callback, guardrail, plugin, sandbox, or provider. Those
+				// boundaries are not all cooperative and must not regain withdrawn
+				// authority merely because the turn record still had to be created.
+				ctx.abortController.signal.throwIfAborted()
+
+				// Handed over here, and the position is load-bearing in three
+				// directions. It has to follow `wirePlanManager`, or a host that
+				// builds its plan in this callback does it into silence. It has to
+				// follow `recorder.init()` and `turn_started`, because plan events append
+				// to that durable turn. It also has to follow the pre-model abort fence:
+				// a callback invoked after attachment resolution observed cancellation
+				// would regain withdrawn authority and could replace the cancellation
+				// with its own failure. This is still before the iteration loop, which
+				// is the guarantee the callback makes.
+				params.onContextCreated?.({ planManager: ctx.planManager })
+
+				// The box is handed out HERE, after `turn_started`, and the position
+				// is load-bearing rather than tidy. It moved twice:
+				//
+				//  1. Beside the box's construction — a host that called `set`
+				//     synchronously reached `eventTranslator` inside its temporal
+				//     dead zone and killed the turn before it started.
+				//  2. Beside the translator's construction — the translator existed,
+				//     but the turn directory did not, so the durable append hit
+				//     ENOENT on `transcript.jsonl`.
+				//
+				// Both were found by the test that takes the box and immediately
+				// swaps the policy, which is not an exotic host: it is the shape of
+				// "start unattended" wiring. A policy change is durably recorded
+				// before it takes effect, so the handout cannot precede the turn
+				// being writable.
+				params.onApprovalPolicy?.(approvalPolicy)
+
+				// History repair happens before the turn manager sees the first model
+				// request, but its durable event cannot precede turn_started: there is no
+				// writable run log until that event initializes it. Emit the measured
+				// counts here, still before any provider call, so hosts can tell that the
+				// model received a repaired projection rather than the raw history.
+				for (const repair of pendingHistoryRepairs) {
+					await eventTranslator.emitEvent({
+						type: 'message_history_repaired',
+						turnId: ctx.recorder.turnId,
+						source: repair.source,
+						...repair.report,
+					})
+					yield* eventTranslator.drainPending()
+				}
 
-			// --- Sandbox lifecycle: create before iteration loop ---
-			if (params.sandboxProvider) {
-				const rootAtCwd = runConfig.sandbox?.workspace === 'working-directory'
-				// Checked against what the CALLER passed, not against `ctx.cwd`.
-				// `ctx.cwd` falls back to `process.cwd()`, so reading it here
-				// would silently root the sandbox at whatever directory the
-				// host process happens to be in — which is not the directory
-				// anybody asked to confine, and is worse than the temp dir the
-				// caller declined. Refused before the sandbox exists, and not
-				// downgraded to ephemeral: a caller who asked for confinement
-				// of a specific tree and quietly got an empty one has been
-				// told their files are protected by something that is not
-				// looking at them.
-				if (rootAtCwd && params.workingDirectory === undefined) {
-					throw new NamzuError({
-						code: 'invalid_config',
-						message:
-							"sandbox.workspace is 'working-directory' but this run has no workingDirectory. Pass one, or use the default 'ephemeral' — the kernel will not fall back to a temp directory, because that would confine a directory you did not name.",
-						details: { workspace: 'working-directory' },
+				// Surface capability degradation to the host as run events —
+				// explicit, not silent (the log.warn above fires at setup time;
+				// this is the machine-readable channel).
+				if (stripToolSurfaces) {
+					await eventTranslator.emitEvent({
+						type: 'capability_warning',
+						turnId: ctx.recorder.turnId,
+						capability: 'tools',
+						providerId: params.provider.id,
+						message: `Provider '${params.provider.id}' does not support tools — ${registeredToolCount} registered tool(s) were stripped from the prompt and request.`,
 					})
+					yield* eventTranslator.drainPending()
 				}
-				if (rootAtCwd && !params.sandboxProvider.workspaceModes?.includes('working-directory')) {
-					throw new NamzuError({
-						code: 'invalid_config',
-						message: `Sandbox provider '${params.sandboxProvider.id}' does not advertise working-directory workspace support. Refusing instead of passing a project path the provider may ignore.`,
-						details: {
-							providerId: params.sandboxProvider.id,
-							workspace: 'working-directory',
-						},
+				if (attachmentMessageCount > 0) {
+					await eventTranslator.emitEvent({
+						type: 'capability_warning',
+						turnId: ctx.recorder.turnId,
+						capability: 'vision',
+						providerId: params.provider.id,
+						message: `Provider '${params.provider.id}' does not support vision — image attachments on ${attachmentMessageCount} user message(s) will not reach the model.`,
 					})
+					yield* eventTranslator.drainPending()
 				}
-				const acquisition = await acquireSandbox({
-					provider: params.sandboxProvider,
-					config: {
-						...(rootAtCwd ? { workingDirectory: ctx.cwd } : {}),
-						...(rootAtCwd && params.additionalDirectories?.length
-							? { additionalDirectories: params.additionalDirectories }
-							: {}),
-						timeoutMs: runConfig.sandbox?.timeoutMs,
-						memoryLimitMb: runConfig.sandbox?.memoryLimitMb,
-						maxProcesses: runConfig.sandbox?.maxProcesses,
-					},
-					signal: ctx.abortController.signal,
-					timeoutMs: guard.remainingUntilTimeoutMs(),
-					teardownTimeoutMs: sandboxTeardownTimeoutMs,
-					onLateTeardown: (result) => {
-						if (result.kind === 'destroyed') {
-							ctx.log.info('Late sandbox allocation was destroyed after the run stopped', {
-								[NAMZU.RUN_ID]: ctx.runId,
-							})
-							return
-						}
-						ctx.log.error('Late sandbox allocation did not stop cleanly', {
-							[NAMZU.RUN_ID]: ctx.runId,
-							...errorAttributes(result.error),
+				if (documentMessageCount > 0) {
+					await eventTranslator.emitEvent({
+						type: 'capability_warning',
+						turnId: ctx.recorder.turnId,
+						capability: 'documents',
+						providerId: params.provider.id,
+						message: `Provider '${params.provider.id}' does not support documents — document attachments on ${documentMessageCount} user message(s) will not reach the model.`,
+					})
+					yield* eventTranslator.drainPending()
+				}
+
+				if (params.pluginManager) {
+					const hookResults = await params.pluginManager.executeHooks(
+						'turn_start',
+						{ sessionId: ctx.sessionId, turnId: ctx.turnId, signal: ctx.abortController.signal },
+						eventTranslator.emitEvent,
+					)
+					applyLifecycleHookResults('turn_start', hookResults)
+					yield* eventTranslator.drainPending()
+				}
+
+				// --- Sandbox lifecycle: create before iteration loop ---
+				if (params.sandboxProvider) {
+					const rootAtCwd = turnConfig.sandbox?.workspace === 'working-directory'
+					// Checked against what the CALLER passed, not against `ctx.cwd`.
+					// `ctx.cwd` falls back to `process.cwd()`, so reading it here
+					// would silently root the sandbox at whatever directory the
+					// host process happens to be in — which is not the directory
+					// anybody asked to confine, and is worse than the temp dir the
+					// caller declined. Refused before the sandbox exists, and not
+					// downgraded to ephemeral: a caller who asked for confinement
+					// of a specific tree and quietly got an empty one has been
+					// told their files are protected by something that is not
+					// looking at them.
+					if (rootAtCwd && params.workingDirectory === undefined) {
+						throw new NamzuError({
+							code: 'invalid_config',
+							message:
+								"sandbox.workspace is 'working-directory' but this turn has no workingDirectory. Pass one, or use the default 'ephemeral' — the kernel will not fall back to a temp directory, because that would confine a directory you did not name.",
+							details: { workspace: 'working-directory' },
 						})
-					},
-				})
-				if (acquisition.kind !== 'created') {
-					if (acquisition.createPending) {
-						ctx.log.warn(
-							'Sandbox creation was unsettled when the run stopped; any returned handle will be released, but a remote allocation hidden behind a lost response requires provider-side reconciliation or a fleet reaper',
-							{
-								[NAMZU.RUN_ID]: ctx.runId,
-								'namzu.sandbox.provider_id': params.sandboxProvider.id,
-							},
-						)
 					}
-					if (acquisition.kind === 'cancelled') {
-						ctx.runMgr.markCancelled()
-					} else {
-						ctx.runMgr.setStopReason('timeout')
-						ctx.log.warn('Sandbox creation exhausted the run timeout', {
-							[NAMZU.RUN_ID]: ctx.runId,
-							'namzu.sandbox.provider_id': params.sandboxProvider.id,
-							...errorAttributes(acquisition.error),
+					if (rootAtCwd && !params.sandboxProvider.workspaceModes?.includes('working-directory')) {
+						throw new NamzuError({
+							code: 'invalid_config',
+							message: `Sandbox provider '${params.sandboxProvider.id}' does not advertise working-directory workspace support. Refusing instead of passing a project path the provider may ignore.`,
+							details: {
+								providerId: params.sandboxProvider.id,
+								workspace: 'working-directory',
+							},
 						})
 					}
-					yield* resultAssembler.completeRun(rootSpan)
-					// The run HAS settled, so the outer `finally` must not read
-					// this as an abandonment — it would persist a second time.
+					const acquisition = await acquireSandbox({
+						provider: params.sandboxProvider,
+						config: {
+							...(rootAtCwd ? { workingDirectory: ctx.cwd } : {}),
+							...(rootAtCwd && params.additionalDirectories?.length
+								? { additionalDirectories: params.additionalDirectories }
+								: {}),
+							timeoutMs: turnConfig.sandbox?.timeoutMs,
+							memoryLimitMb: turnConfig.sandbox?.memoryLimitMb,
+							maxProcesses: turnConfig.sandbox?.maxProcesses,
+						},
+						signal: ctx.abortController.signal,
+						timeoutMs: guard.remainingUntilTimeoutMs(),
+						teardownTimeoutMs: sandboxTeardownTimeoutMs,
+						onLateTeardown: (result) => {
+							if (result.kind === 'destroyed') {
+								ctx.log.info('Late sandbox allocation was destroyed after the turn stopped', {
+									[NAMZU.TURN_ID]: ctx.turnId,
+								})
+								return
+							}
+							ctx.log.error('Late sandbox allocation did not stop cleanly', {
+								[NAMZU.TURN_ID]: ctx.turnId,
+								...errorAttributes(result.error),
+							})
+						},
+					})
+					if (acquisition.kind !== 'created') {
+						if (acquisition.createPending) {
+							ctx.log.warn(
+								'Sandbox creation was unsettled when the turn stopped; any returned handle will be released, but a remote allocation hidden behind a lost response requires provider-side reconciliation or a fleet reaper',
+								{
+									[NAMZU.TURN_ID]: ctx.turnId,
+									'namzu.sandbox.provider_id': params.sandboxProvider.id,
+								},
+							)
+						}
+						if (acquisition.kind === 'cancelled') {
+							ctx.recorder.markCancelled()
+						} else {
+							ctx.recorder.setStopReason('timeout')
+							ctx.log.warn('Sandbox creation exhausted the turn timeout', {
+								[NAMZU.TURN_ID]: ctx.turnId,
+								'namzu.sandbox.provider_id': params.sandboxProvider.id,
+								...errorAttributes(acquisition.error),
+							})
+						}
+						yield* resultAssembler.completeTurn(rootSpan)
+						// The turn HAS settled, so the outer `finally` must not read
+						// this as an abandonment — it would persist a second time.
+						settled = true
+						return await resultAssembler.finalize()
+					}
+					sandbox = acquisition.sandbox
+					toolExecutor.setSandbox(sandbox)
+
+					await eventTranslator.emitEvent({
+						type: 'sandbox_created',
+						turnId: ctx.turnId,
+						sandboxId: sandbox.id,
+						environment: sandbox.environment,
+					})
+					yield* eventTranslator.drainPending()
+
+					ctx.log.info('Sandbox created for run', {
+						'namzu.sandbox.id': sandbox.id,
+						'namzu.execution.environment': sandbox.environment,
+						'namzu.runtime.root_dir': sandbox.rootDir,
+					})
+				}
+
+				// Before the first model call: the cheapest place to refuse, since
+				// nothing has been spent. Previously unreachable — `turn_start`
+				// fires with only `{ sessionId, turnId }` and `turn_started` carries only the
+				// system prompt, so no hook could see the user's message.
+				const inputVerdict = await runInputGuardrails(
+					params.inputGuardrails,
+					{
+						sessionId: ctx.sessionId,
+						turnId: ctx.turnId,
+						messages: ctx.recorder.messages,
+						...(segments?.static ? { systemPrompt: segments.static } : {}),
+					},
+					ctx.log,
+				)
+				if (inputVerdict.blocked) {
+					await eventTranslator.emitEvent({
+						type: 'guardrail_triggered',
+						turnId: ctx.turnId,
+						stage: 'input',
+						action: 'block',
+						...(inputVerdict.name ? { guardrail: inputVerdict.name } : {}),
+						...(inputVerdict.reason ? { reason: inputVerdict.reason } : {}),
+					})
+					yield* eventTranslator.drainPending()
+					// A guardrail block is a refusal — first-class in the audit trail
+					// (LOG-14, design §5), not merely a SessionEvent a host happens to be
+					// subscribed to when it fires.
+					await ctx.recorder.recordAudit({
+						what: { action: 'guardrail:input', resource: inputVerdict.name },
+						outcome: 'refused',
+						reason: inputVerdict.reason ?? 'blocked by an input guardrail',
+						...(params.persona?.identity.role ? { persona: params.persona.identity.role } : {}),
+					})
+					ctx.recorder.setStopReason('input_guardrail')
+					ctx.recorder.setLastError(inputVerdict.reason ?? 'blocked by an input guardrail')
+					yield* resultAssembler.completeTurn(rootSpan)
+					// Same two lines as the sandbox path above, and for the same
+					// reasons — with one that path does not have. This return used to
+					// hand back `getRun()` without persisting, so the terminal state
+					// reached the disk only because the abandonment path found
+					// `settled` false and settled it a second time. A branch that
+					// exists for runs which did NOT settle must not be the reason a
+					// settled one is written down.
 					settled = true
 					return await resultAssembler.finalize()
 				}
-				sandbox = acquisition.sandbox
-				toolExecutor.setSandbox(sandbox)
-
-				await eventTranslator.emitEvent({
-					type: 'sandbox_created',
-					runId: ctx.runId,
-					sandboxId: sandbox.id,
-					environment: sandbox.environment,
-				})
-				yield* eventTranslator.drainPending()
 
-				ctx.log.info('Sandbox created for run', {
-					'namzu.sandbox.id': sandbox.id,
-					'namzu.execution.environment': sandbox.environment,
-					'namzu.runtime.root_dir': sandbox.rootDir,
-				})
-			}
+				// Honor the approval a human already gave, before the loop's
+				// first model call. The sandbox exists by now, so an approved
+				// tool that needs one gets it.
+				if (pendingResume) {
+					ctx.log.info('Applying a pending HITL decision to the checkpointed tool calls', {
+						[NAMZU.TURN_ID]: ctx.turnId,
+						'namzu.tool.names': pendingResume.response.message.toolCalls?.map(
+							(tc) => tc.function.name,
+						),
+						'namzu.runtime.denied': pendingResume.denials.size,
+					})
+					// Hand the recorded answer to the already-built tool. The tool
+					// closed over its registry when the agent was constructed,
+					// long before this turn existed, so the answers are copied in
+					// rather than passed down.
+					if (pendingResume.answers) {
+						for (const [questionId, answer] of pendingResume.answers.entries()) {
+							pendingAnswers.set(questionId, answer)
+						}
+					}
 
-			// Crash-save handlers live for exactly the run's lifetime, and are
-			// removed in the `finally` below. Opt-in — see `emergencySave`.
-			if (params.emergencySave) {
-				const runDir = ctx.runMgr.getRunDir()
-				if (runDir) {
-					emergencyManager = EmergencySaveManager.instance(ctx.log)
-					emergencyManager.attach(ctx.runMgr, runDir, ctx.log)
-				} else {
-					ctx.log.warn(
-						'emergencySave requested but the run has no output directory — crash dumps disabled',
-						{ [NAMZU.RUN_ID]: ctx.runId },
+					await applyPendingResume(
+						pendingResume,
+						ctx.recorder,
+						toolExecutor,
+						recoveredResults,
+						pendingResumeAssistantId,
 					)
+					yield* eventTranslator.drainPending()
 				}
-			}
 
-			// Before the first model call: the cheapest place to refuse, since
-			// nothing has been spent. Previously unreachable — `run_start`
-			// fires with only `{ runId }` and `run_started` carries only the
-			// system prompt, so no hook could see the user's message.
-			const inputVerdict = await runInputGuardrails(
-				params.inputGuardrails,
-				{
-					runId: ctx.runId,
-					messages: ctx.runMgr.messages,
-					...(segments?.static ? { systemPrompt: segments.static } : {}),
-				},
-				ctx.log,
-			)
-			if (inputVerdict.blocked) {
-				await eventTranslator.emitEvent({
-					type: 'guardrail_triggered',
-					runId: ctx.runId,
-					stage: 'input',
-					action: 'block',
-					...(inputVerdict.name ? { guardrail: inputVerdict.name } : {}),
-					...(inputVerdict.reason ? { reason: inputVerdict.reason } : {}),
-				})
-				yield* eventTranslator.drainPending()
-				// A guardrail block is a refusal — first-class in the audit trail
-				// (LOG-14, design §5), not merely a RunEvent a host happens to be
-				// subscribed to when it fires.
-				await ctx.runMgr.recordAudit({
-					what: { action: 'guardrail:input', resource: inputVerdict.name },
-					outcome: 'refused',
-					reason: inputVerdict.reason ?? 'blocked by an input guardrail',
-					...(params.persona?.identity.role ? { persona: params.persona.identity.role } : {}),
-				})
-				ctx.runMgr.setStopReason('input_guardrail')
-				ctx.runMgr.setLastError(inputVerdict.reason ?? 'blocked by an input guardrail')
-				yield* resultAssembler.completeRun(rootSpan)
-				// Same two lines as the sandbox path above, and for the same
-				// reasons — with one that path does not have. This return used to
-				// hand back `getRun()` without persisting, so the terminal state
-				// reached the disk only because the abandonment path found
-				// `settled` false and settled it a second time. A branch that
-				// exists for runs which did NOT settle must not be the reason a
-				// settled one is written down.
-				settled = true
-				return await resultAssembler.finalize()
-			}
-
-			// Honor the approval a human already gave, before the loop's
-			// first model call. The sandbox exists by now, so an approved
-			// tool that needs one gets it.
-			if (pendingResume) {
-				ctx.log.info('Applying a pending HITL decision to the checkpointed tool calls', {
-					[NAMZU.RUN_ID]: ctx.runId,
-					'namzu.tool.names': pendingResume.response.message.toolCalls?.map(
-						(tc) => tc.function.name,
-					),
-					'namzu.runtime.denied': pendingResume.denials.size,
-				})
-				// Hand the recorded answer to the already-built tool. The tool
-				// closed over its registry when the agent was constructed,
-				// long before this run existed, so the answers are copied in
-				// rather than passed down.
-				if (pendingResume.answers) {
-					for (const [questionId, answer] of pendingResume.answers.entries()) {
-						pendingAnswers.set(questionId, answer)
-					}
+				// The decision has now actually been carried out, so the park it
+				// answered is no longer outstanding. Without this the checkpoint
+				// keeps reporting `pending` with no `resolvedAt`, and an approval
+				// queue re-serves a call that already ran — or a question already
+				// answered — which defeats the entire point of recording the park.
+				//
+				// Two arms reach this point, and being outside `if (pendingResume)`
+				// is what the second one needs. One is a plan whose decision was
+				// applied to a batch above. The other is the cadence arm, for which
+				// `planPendingResume` rightly produces no plan because the loop
+				// resuming IS its decision being carried out (`answeredParkId`, set
+				// on the restore path). Resolving only the first left a finished turn
+				// reporting `awaiting-decision` forever.
+				//
+				// What is RECORDED depends on which of the two produced the plan. A
+				// recovery plan means the batch was answered by the crash path
+				// rather than by the decision — the calls the human was asked about
+				// were closed with explicitly unknown outcomes — so the human's
+				// answer must not be written down as what ended the park. The park is
+				// still resolved: the question is moot, and leaving it outstanding
+				// would have `findPendingCheckpoint` serve it as the newest
+				// outstanding park, so a host resuming it would rewind this turn to
+				// the checkpoint the crash happened on and re-execute a batch the turn
+				// has long since moved past.
+				const resolvedCheckpointId = pendingResume?.checkpointId ?? answeredParkId
+				const recordedDecision =
+					pendingResume?.source === 'recovery' && params.pendingDecision
+						? supersededByRecovery(params.pendingDecision)
+						: params.pendingDecision
+				if (recordedDecision && resolvedCheckpointId) {
+					await checkpointMgr
+						.unpark(resolvedCheckpointId, recordedDecision)
+						.catch((err: unknown) => {
+							ctx.log.error('Applied a pending decision but failed to clear the park', {
+								[NAMZU.TURN_ID]: ctx.turnId,
+								'namzu.checkpoint.id': resolvedCheckpointId,
+								'exception.message': err instanceof Error ? err.message : String(err),
+							})
+							return null
+						})
 				}
 
-				await applyPendingResume(pendingResume, ctx.runMgr, toolExecutor, recoveredResults)
-				yield* eventTranslator.drainPending()
-			}
+				yield* iterationOrchestrator.runLoop()
 
-			// The decision has now actually been carried out, so the park it
-			// answered is no longer outstanding. Without this the checkpoint
-			// keeps reporting `pending` with no `resolvedAt`, and an approval
-			// queue re-serves a call that already ran — or a question already
-			// answered — which defeats the entire point of recording the park.
-			//
-			// Two arms reach this point, and being outside `if (pendingResume)`
-			// is what the second one needs. One is a plan whose decision was
-			// applied to a batch above. The other is the cadence arm, for which
-			// `planPendingResume` rightly produces no plan because the loop
-			// resuming IS its decision being carried out (`answeredParkId`, set
-			// on the restore path). Resolving only the first left a finished run
-			// reporting `awaiting-decision` forever.
-			//
-			// What is RECORDED depends on which of the two produced the plan. A
-			// recovery plan means the batch was answered by the crash path
-			// rather than by the decision — the calls the human was asked about
-			// were closed with explicitly unknown outcomes — so the human's
-			// answer must not be written down as what ended the park. The park is
-			// still resolved: the question is moot, and leaving it outstanding
-			// would have `findPendingCheckpoint` serve it as the newest
-			// outstanding park, so a host resuming it would rewind this run to
-			// the checkpoint the crash happened on and re-execute a batch the run
-			// has long since moved past.
-			const resolvedCheckpointId = pendingResume?.checkpointId ?? answeredParkId
-			const recordedDecision =
-				pendingResume?.source === 'recovery' && params.pendingDecision
-					? supersededByRecovery(params.pendingDecision)
-					: params.pendingDecision
-			if (recordedDecision && resolvedCheckpointId) {
-				await checkpointMgr.unpark(resolvedCheckpointId, recordedDecision).catch((err: unknown) => {
-					ctx.log.error('Applied a pending decision but failed to clear the park', {
-						[NAMZU.RUN_ID]: ctx.runId,
-						'namzu.checkpoint.id': resolvedCheckpointId,
-						'exception.message': err instanceof Error ? err.message : String(err),
-					})
-					return null
+				yield* finalizeTurn({
+					ctx,
+					params,
+					eventTranslator,
+					takeSteps: () => iterationOrchestrator.getSteps(),
+					workingStateManager,
+					resultAssembler,
+					rootSpan,
+				})
+			} catch (err) {
+				// Another turn is active in this session: nothing of this turn was
+				// written, so there is nothing to settle. The caller gets the
+				// refusal itself.
+				if (isTurnInProgressError(err) && !ctx.recorder.isActive) throw err
+				// A failed turn still spent its steps; report them.
+				ctx.recorder.setSteps(iterationOrchestrator.getSteps())
+				await executeUserInterruptHooks(err)
+				yield* eventTranslator.drainPending()
+				yield* resultAssembler.handleError(err, rootSpan)
+			} finally {
+				yield* releaseTurnResources({
+					ctx,
+					eventTranslator,
+					unsubscribeJobExits,
+					unsubscribeTaskStore,
+					awaitedJobs,
+					backgroundJobs: params.backgroundJobs,
+					backgroundJobOwner: params.backgroundJobOwner,
+					questionParks,
+					workingStateManager,
+					promoteMemory: params.promoteMemory,
+					sandbox,
+					sandboxTeardownTimeoutMs,
+					runStartedAt,
+					rootSpan,
 				})
 			}
 
-			yield* iterationOrchestrator.runLoop()
+			// Reached only by a turn that settled on its own terms. `finalize()` is
+			// the only thing in this body that writes the durable half of the turn,
+			// and a `return` completion arriving from a consumer (`break` out of
+			// `for await`, `gen.return()`) runs the `finally` above and stops short
+			// of here. The flag is what tells the two apart, and this is one of
+			// three sites that set it — the sandbox-acquisition and input-guardrail
+			// returns settle early and set it there. Set before the await rather
+			// than after, because a store that throws on the way out must not send
+			// the abandonment path over the same broken ground.
+			settled = true
+			return await resultAssembler.finalize()
+		})()
 
-			yield* finalizeRun({
-				ctx,
-				params,
-				eventTranslator,
-				takeSteps: () => iterationOrchestrator.getSteps(),
-				workingStateManager,
-				resultAssembler,
-				rootSpan,
-			})
-		} catch (err) {
-			// A failed run still spent its steps; report them.
-			ctx.runMgr.setSteps(iterationOrchestrator.getSteps())
-			await executeUserInterruptHooks(err)
-			yield* eventTranslator.drainPending()
-			yield* resultAssembler.handleError(err, rootSpan)
+		try {
+			return yield* runBody
 		} finally {
-			yield* releaseRunResources({
-				ctx,
-				eventTranslator,
-				emergencyManager,
-				unsubscribeJobExits,
-				unsubscribeTaskStore,
-				awaitedJobs,
-				backgroundJobs: params.backgroundJobs,
-				backgroundJobOwner: params.backgroundJobOwner,
-				questionParks,
-				workingStateManager,
-				promoteMemory: params.promoteMemory,
-				sandbox,
-				sandboxTeardownTimeoutMs,
-				runStartedAt,
-				rootSpan,
-			})
+			if (!settled) await settleAbandonedTurn(ctx.recorder, eventTranslator, ctx.log)
 		}
-
-		// Reached only by a run that settled on its own terms. `finalize()` is
-		// the only thing in this body that writes the durable half of the run,
-		// and a `return` completion arriving from a consumer (`break` out of
-		// `for await`, `gen.return()`) runs the `finally` above and stops short
-		// of here. The flag is what tells the two apart, and this is one of
-		// three sites that set it — the sandbox-acquisition and input-guardrail
-		// returns settle early and set it there. Set before the await rather
-		// than after, because a store that throws on the way out must not send
-		// the abandonment path over the same broken ground.
-		settled = true
-		return await resultAssembler.finalize()
-	})()
-
-	try {
-		return yield* runBody
 	} finally {
-		if (!settled) await settleAbandonedRun(ctx.runMgr, ctx.log)
+		unsubscribeChildSessions?.()
+		// The lease `open()` took in the prelude; a caller-supplied lease is the
+		// caller's to give back.
+		await ctx.recorder.release()
 	}
 }
 
 /**
- * Write a terminal durable record for a run whose consumer walked away.
- *
- * `for await (… ) break` and an explicit `gen.return()` both end the run
- * body early. Everything the run's `finally` owns still happens — background
- * jobs are killed, the sandbox is destroyed, the span ends, the duration is
- * recorded — and then the generator stops. `finalize()` never runs, so
- * `persist()` never runs, and the store keeps whatever `init()` wrote: a
- * non-terminal status for a run that no longer exists. `deriveRunStatus`
- * reads that record back as `queued`, work waiting to start, and a host
- * rebuilding its view from the store believes it.
+ * Write a durable record for a turn whose consumer walked away.
  *
- * There is nothing to emit here and nothing to emit it to: the consumer
- * that would have received the events is the one that left. This is about
- * the durable record only.
+ * `for await (… ) break` and an explicit `gen.return()` both end the turn
+ * body early. Everything the body's `finally` owns still happens — jobs are
+ * killed, the sandbox is destroyed, the span ends — and then the generator
+ * stops without settling. Without this the log would end mid-turn, and the
+ * turn would read as `interrupted` for a process that is gone.
  *
- * `cancelled` is the verdict, and it is chosen from the existing vocabulary
- * because it is the one that is true. The run did not complete — no result
- * was produced and no terminal event was ever delivered — and nothing
- * failed, so `failed` would name an error that never happened; a run whose
- * consumer stopped reading and whose processes were torn down under it is
- * the same fact `markCancelled` already records when a run abort tears one
- * down. It needs no new `RunExecutionStatus` and no new `StopReason`.
- *
- * A verdict the run already reached is left standing. A run that failed,
- * or was cancelled, before the consumer left still says so; what the
- * abandonment adds is that the record reaches the disk at all.
- *
- * Neither is a verdict written over a PARK. A park is a promise to a human
- * that outlives the consumer: the run is resumable and somebody is still owed
- * an answer, and `deriveRunStatus` reads a terminal status BEFORE it reads the
- * park — so recording `cancelled` turns `awaiting_hitl` into `cancelled` for a
- * run nobody answered for, while the unanswered question stays on the record
- * and the checkpoint it belongs to stays the place a resume starts from. The
- * durable state is asked rather than the in-memory one because the in-memory
- * one is the misleading half here: `handleHITLDecision` emits `run_paused` and
- * drains it BEFORE it calls `setStopReason('paused')`, so a consumer that
- * leaves on that event leaves a run whose status is `running` and whose stop
- * reason is unset at the exact instant its park is already durable.
- * `findPendingCheckpoint` is the same read an approval queue is built from,
- * expired parks included in its judgement: a park nobody answered in time is
- * not somebody still being asked.
+ * - A turn that already settled, or already paused, is left as it stands.
+ * - A turn parked on a decision (an open `decision_requested`) is PAUSED on
+ *   that checkpoint: a park is a promise to a human that outlives the
+ *   consumer, and `resumeSession` continues it once the answer arrives.
+ * - Any other turn is settled `cancelled` (`turn_completed`): it did not
+ *   complete and nothing failed; its consumer stopped reading.
  *
  * Never throws. It runs while an exception may already be unwinding, and a
- * store that cannot be written must not replace the run's real failure with
- * its own.
+ * log that cannot be written must not replace the turn's real failure.
  */
-async function settleAbandonedRun(runMgr: RunPersistence, log: Logger): Promise<void> {
+async function settleAbandonedTurn(
+	recorder: TurnRecorder,
+	eventTranslator: EventTranslator,
+	log: Logger,
+): Promise<void> {
 	try {
-		// A terminal verdict is written whatever the park says: `deriveRunStatus`
-		// settles a run that finished, failed or was cancelled BEFORE it looks at
-		// a park ("terminal beats parked"), so a settled run is not waiting for
-		// anybody and the row it already wrote must reach the disk. This ordering
-		// is also what keeps a stale park from suppressing the write.
-		if (!isTerminalStatus(runMgr.status)) {
-			const parked = await findPendingCheckpoint(runMgr.getCheckpointStore(), runMgr.getRunScope())
-			if (parked) {
-				// Left exactly as it stands: no verdict, no write. The park row is
-				// this run's durable state, and `persist()` here would add a
-				// second claim — `running`, for a process that is gone — beside it.
-				log.info('Abandoned run left parked for a human to answer', {
-					[NAMZU.RUN_ID]: runMgr.id,
-					'namzu.checkpoint.id': parked.id,
-					'namzu.runtime.park_type': parked.pending?.request.type,
-				})
-				return
-			}
-			runMgr.markCancelled()
+		if (recorder.isClosed || recorder.isPaused || !recorder.isActive) return
+		await recorder.flush()
+		const parked = await findPendingCheckpoint(recorder.log, { turnId: recorder.turnId })
+		if (parked) {
+			recorder.setStopReason('paused')
+			await eventTranslator.emitEvent({
+				type: 'turn_paused',
+				checkpointId: parked.checkpointId,
+				reason: 'The consumer stopped reading while the turn was awaiting a decision.',
+			})
+			await recorder.persist()
+			log.info('Abandoned turn left paused for a human to answer', {
+				[NAMZU.TURN_ID]: recorder.turnId,
+				'namzu.checkpoint.id': parked.checkpointId,
+				'namzu.runtime.park_type': parked.pending.request.type,
+			})
+			return
 		}
-		// Once: the `finally` that calls this runs once, and every site in the
-		// run body that settles through `finalize()` sets `settled` before it
-		// returns, so the two can never both write.
-		await runMgr.persist()
-		log.info('Abandoned run recorded as cancelled', {
-			[NAMZU.RUN_ID]: runMgr.id,
+		if (!isTerminalStatus(recorder.status)) recorder.markCancelled()
+		await eventTranslator.emitEvent({
+			type: 'turn_completed',
+			result: recorder.getTurn().result ?? '',
+			...(recorder.stopReason ? { stopReason: recorder.stopReason } : {}),
+			settlement: recorder.settlement(recorder.status === 'cancelled' ? 'cancelled' : 'completed'),
+		})
+		await recorder.persist()
+		log.info('Abandoned turn recorded as cancelled', {
+			[NAMZU.TURN_ID]: recorder.turnId,
 		})
 	} catch (err) {
-		log.error('Failed to record the terminal state of an abandoned run', {
-			[NAMZU.RUN_ID]: runMgr.id,
+		log.error('Failed to record the end of an abandoned turn', {
+			[NAMZU.TURN_ID]: recorder.turnId,
 			'exception.message': err instanceof Error ? err.message : String(err),
 		})
 	}
@@ -2391,8 +2465,8 @@ type DrainQueryParams = Omit<QueryParams, 'resumeHandler'> & {
 
 async function drainPreparedQuery(
 	fullParams: QueryParams,
-	listener?: RunEventListener,
-): Promise<Run> {
+	listener?: SessionEventListener,
+): Promise<Turn> {
 	const gen = query(fullParams)
 	let result = await gen.next()
 
@@ -2410,8 +2484,8 @@ export async function drainQuery(
 	params: Omit<QueryParams, 'resumeHandler'> & {
 		resumeHandler?: ResumeHandler
 	},
-	listener?: RunEventListener,
-): Promise<Run> {
+	listener?: SessionEventListener,
+): Promise<Turn> {
 	const fullParams: QueryParams = {
 		...params,
 		resumeHandler: params.resumeHandler ?? autoApproveHandler,
@@ -2419,12 +2493,12 @@ export async function drainQuery(
 	return await drainPreparedQuery(fullParams, listener)
 }
 
-/** @internal Canonical resume state already selected by `resumeRun`. */
+/** @internal Canonical resume state already selected by `resumeSession`. */
 export async function drainQueryWithSelectedResumeState(
 	params: DrainQueryParams,
 	state: SelectedResumeState,
-	listener?: RunEventListener,
-): Promise<Run> {
+	listener?: SessionEventListener,
+): Promise<Turn> {
 	const fullParams: QueryParams = {
 		...params,
 		resumeHandler: params.resumeHandler ?? autoApproveHandler,
@@ -2439,17 +2513,17 @@ export async function drainQueryWithSelectedResumeState(
 
 function assertSelectedResumeAttribution(params: QueryParams, state: SelectedResumeState): void {
 	const mismatchedFields: string[] = []
-	if (params.runId !== state.runId) mismatchedFields.push('runId')
+	if (params.turnId !== state.turnId) mismatchedFields.push('turnId')
 	if (params.sessionId !== state.sessionId) mismatchedFields.push('sessionId')
 	if (params.topicId !== state.topicId) mismatchedFields.push('topicId')
 	if (params.projectId !== state.projectId) mismatchedFields.push('projectId')
 	if (params.tenantId !== state.tenantId) mismatchedFields.push('tenantId')
-	if (params.parentRunId !== state.parentRunId) mismatchedFields.push('parentRunId')
+	if (params.parentSessionId !== state.parentSessionId) mismatchedFields.push('parentSessionId')
 	if (mismatchedFields.length === 0) return
 
 	throw new NamzuError({
 		code: 'invalid_config',
-		message: 'The selected resume state does not match the run attribution supplied to the query.',
+		message: 'The selected resume state does not match the turn attribution supplied to the query.',
 		details: { fields: mismatchedFields },
 	})
 }