@vellumai/assistant 0.4.55 → 0.4.56

package/ARCHITECTURE.md

@@ -144,23 +144,23 @@ In addition to persistent trust rules (`always_allow` / `always_deny`), the appr
 
 **Two modes:**
 
-1. **`allow_thread`** — Auto-approve all tool confirmations for the remainder of the current conversation. The override persists until the session ends, the conversation is closed, or the mode is explicitly cleared.
+1. **`allow_conversation`** — Auto-approve all tool confirmations for the remainder of the current conversation. The override persists until the session ends, the conversation is closed, or the mode is explicitly cleared.
 2. **`allow_10m`** — Auto-approve all tool confirmations for 10 minutes (configurable). The override expires lazily on the next read after the TTL elapses — no background sweep runs.
 
-**Session-scoped, in-memory only:** Overrides are keyed by `conversationId` and stored in an in-memory `Map` inside `session-approval-overrides.ts`. They do not survive daemon restarts, which is intentional — temporary approvals should not outlive the session that created them.
+**Session-scoped, in-memory only:** Overrides are keyed by `conversationId` and stored in an in-memory `Map` inside `conversation-approval-overrides.ts`. They do not survive daemon restarts, which is intentional — temporary approvals should not outlive the session that created them.
 
 **Integration with the permission pipeline:** The permission checker (`src/tools/permission-checker.ts`) checks for an active temporary override via `getEffectiveMode()` before prompting the user. If an active override exists for the current conversation, the confirmation is auto-approved without surfacing a prompt. This check runs after persistent trust rules, so a persistent `deny` rule still takes precedence.
 
-**No persistent side effects:** Temporary modes do not write to `trust.json` or create persistent trust rules. They are purely ephemeral. The `buildDecisionActions()` function in `guardian-decision-types.ts` controls whether temporary options (`allow_10m`, `allow_thread`) are surfaced in the approval prompt UI, gated by the `temporaryOptionsAvailable` flag.
+**No persistent side effects:** Temporary modes do not write to `trust.json` or create persistent trust rules. They are purely ephemeral. The `buildDecisionActions()` function in `guardian-decision-types.ts` controls whether temporary options (`allow_10m`, `allow_conversation`) are surfaced in the approval prompt UI, gated by the `temporaryOptionsAvailable` flag.
 
 **Key source files:**
 
-| File                                        | Purpose                                                                                                            |
-| ------------------------------------------- | ------------------------------------------------------------------------------------------------------------------ |
-| `src/runtime/session-approval-overrides.ts` | In-memory store: `setThreadMode`, `setTimedMode`, `getEffectiveMode`, `clearMode`, `hasActiveOverride`, `clearAll` |
-| `src/permissions/types.ts`                  | `UserDecision` type (includes `allow_10m`, `allow_thread`, `temporary_override`), `isAllowDecision()` helper       |
-| `src/runtime/guardian-decision-types.ts`    | `buildDecisionActions()` — controls which temporary options appear in approval prompts                             |
-| `src/tools/permission-checker.ts`           | Permission pipeline integration — checks temporary overrides before prompting                                      |
+| File                                             | Purpose                                                                                                                  |
+| ------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------ |
+| `src/runtime/conversation-approval-overrides.ts` | In-memory store: `setConversationMode`, `setTimedMode`, `getEffectiveMode`, `clearMode`, `hasActiveOverride`, `clearAll` |
+| `src/permissions/types.ts`                       | `UserDecision` type (includes `allow_10m`, `allow_conversation`, `temporary_override`), `isAllowDecision()` helper       |
+| `src/runtime/guardian-decision-types.ts`         | `buildDecisionActions()` — controls which temporary options appear in approval prompts                                   |
+| `src/tools/permission-checker.ts`                | Permission pipeline integration — checks temporary overrides before prompting                                            |
 
 ### Canonical Guardian Request System
 
@@ -176,7 +176,7 @@ The canonical guardian request system provides a channel-agnostic, unified domai
 
 4. **Deterministic API (prompt listing and decision endpoints):** Desktop clients and API consumers use `GET /v1/guardian-actions/pending` and `POST /v1/guardian-actions/decision` (HTTP). These endpoints surface canonical requests alongside legacy pending interactions and channel approval records, with deduplication to avoid double-rendering.
 
-5. **Buttons first, text fallback:** All request kinds (`tool_approval`, `pending_question`, `access_request`) are rendered as structured button cards when displayed in macOS/iOS guardian threads. Each prompt also embeds deterministic text fallback instructions (request-code-based approve/reject directives, and for `access_request` the "open invite flow" phrase) so text-based channels and manual fallback always work. Code-only messages (just a request code without decision text) return clarification instead of auto-approving. Disambiguation with multiple pending requests stays fail-closed — no auto-resolve when the target is ambiguous.
+5. **Buttons first, text fallback:** All request kinds (`tool_approval`, `pending_question`, `access_request`) are rendered as structured button cards when displayed in macOS/iOS guardian conversations. Each prompt also embeds deterministic text fallback instructions (request-code-based approve/reject directives, and for `access_request` the "open invite flow" phrase) so text-based channels and manual fallback always work. Code-only messages (just a request code without decision text) return clarification instead of auto-approving. Disambiguation with multiple pending requests stays fail-closed — no auto-resolve when the target is ambiguous.
 
 **Resolver registry:** Kind-specific resolvers (`src/approvals/guardian-request-resolvers.ts`) handle side effects after CAS resolution. Built-in resolvers: `tool_approval` (channel/desktop approval path), `pending_question` (voice call question path), and `access_request` (trusted-contact verification session creation). New request kinds register resolvers without touching the core primitive.
 
@@ -273,7 +273,7 @@ When a voice call's ASK_GUARDIAN consultation times out before the guardian resp
  [completed] or [failed] (terminal)
 ```
 
-**Generated messaging requirement:** All user-facing copy in the guardian timeout/follow-up path is generated through the `guardian-action-message-composer.ts` composition system, which uses a 2-tier priority chain: (1) daemon-injected LLM generator for natural, varied text; (2) deterministic fallback templates for reliability. No hardcoded user-facing strings exist in the flow files (call-controller, inbound-message-handler, session-process) outside of internal log messages and LLM-instruction prompts. A guard test (`guardian-action-no-hardcoded-copy.test.ts`) enforces this invariant.
+**Generated messaging requirement:** All user-facing copy in the guardian timeout/follow-up path is generated through the `guardian-action-message-composer.ts` composition system, which uses a 2-tier priority chain: (1) daemon-injected LLM generator for natural, varied text; (2) deterministic fallback templates for reliability. No hardcoded user-facing strings exist in the flow files (call-controller, inbound-message-handler, conversation-process) outside of internal log messages and LLM-instruction prompts. A guard test (`guardian-action-no-hardcoded-copy.test.ts`) enforces this invariant.
 
 **Callback branch:** When the conversation engine classifies the guardian's intent as `call_back`, the executor starts an outbound call to the counterparty with context about the guardian's answer. The counterparty phone number is resolved from the original call session by call direction (inbound: `fromNumber`; outbound: `toNumber`).
 
@@ -288,7 +288,7 @@ When a voice call's ASK_GUARDIAN consultation times out before the guardian resp
 | `src/daemon/guardian-action-generators.ts`              | Daemon-injected generator factories: `createGuardianActionCopyGenerator` (latency-optimized text rewriting) and `createGuardianFollowUpConversationGenerator` (tool-calling intent classification) |
 | `src/calls/call-controller.ts`                          | Voice timeout handling: marks requests as timed out, sends expiry notices, injects `[GUARDIAN_TIMEOUT]` instruction for generated voice response                                                   |
 | `src/runtime/routes/inbound-message-handler.ts`         | Late reply interception for Telegram channels: matches late answers to expired requests, routes follow-up conversation turns, dispatches actions                                                   |
-| `src/daemon/session-process.ts`                         | Late reply interception for mac channel: same logic as inbound-message-handler but using conversation-ID-based delivery lookup                                                                     |
+| `src/daemon/conversation-process.ts`                    | Late reply interception for mac channel: same logic as inbound-message-handler but using conversation-ID-based delivery lookup                                                                     |
 | `src/calls/guardian-action-sweep.ts`                    | Periodic sweep for stale pending requests; sends expiry notices to guardian destinations                                                                                                           |
 | `src/memory/migrations/030-guardian-action-followup.ts` | Schema migration adding follow-up columns (`followup_state`, `late_answer_text`, `late_answered_at`, `followup_action`, `followup_completed_at`)                                                   |
 
@@ -617,7 +617,7 @@ Release-driven update notification system that surfaces release notes to the ass
 
 The assistant feature-flag resolver (`src/config/assistant-feature-flags.ts`) is the canonical module for determining whether an assistant feature flag is enabled. It loads default values from the unified registry at `meta/feature-flags/feature-flag-registry.json` (bundled copy at `src/config/feature-flag-registry.json`) and resolves the effective state for each declared assistant-scope flag. Assistant feature flags are declaration-driven assistant-scoped booleans that can gate any assistant behavior; skill availability is one consumer.
 
-**Canonical key format:** `feature_flags.<flag_id>.enabled` (e.g., `feature_flags.hatch-new-assistant.enabled`).
+**Canonical key format:** `feature_flags.<flag_id>.enabled` (e.g., `feature_flags.contacts.enabled`).
 
 **Resolution priority** (highest wins):
 
@@ -634,14 +634,14 @@ The assistant feature-flag resolver (`src/config/assistant-feature-flags.ts`) is
 
 **Skill-gating guarantee:** Skill feature-flag gating is **opt-in**: only skills whose SKILL.md frontmatter contains a `featureFlag` field are gated. Skills without the field are always available regardless of feature flag state. For skills that declare a `featureFlag`, when the corresponding flag is OFF the skill is unavailable everywhere — it cannot appear in client UIs, model context, or runtime tool execution. This is enforced at six independent points:
 
-| Enforcement Point                  | Module                                                   | Effect                                                                                                                                                                                                      |
-| ---------------------------------- | -------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| **1. Client skill list**           | `resolveSkillStates()` in `config/skill-state.ts`        | Skills with flag OFF are excluded from the resolved list returned to clients (macOS skill list, settings UI). The skill never appears in the client.                                                        |
-| **2. System prompt skill catalog** | `appendSkillsCatalog()` in `prompts/system-prompt.ts`    | The model-visible `## Skills Catalog` section in the system prompt filters out flagged-off skills. The model cannot see or reference them.                                                                  |
-| **3. `skill_load` tool**           | `executeSkillLoad()` in `tools/skills/load.ts`           | If the model attempts to load a flagged-off skill by name, the tool returns an error: `"skill is currently unavailable (disabled by feature flag)"`.                                                        |
-| **4. Runtime tool projection**     | `projectSkillTools()` in `daemon/session-skill-tools.ts` | Even if a skill was previously active in a session (has `<loaded_skill>` markers in history), the per-turn projection drops it when the flag is OFF. Already-registered tools are unregistered.             |
-| **5. Included child skills**       | `executeSkillLoad()` in `tools/skills/load.ts`           | When a parent skill includes children via the `includes` directive, each child is independently checked against its feature flag. Flagged-off children are silently excluded from the loaded skill content. |
-| **6. Skill install gate**          | `installSkill()` in `daemon/handlers/skills.ts`           | When a client requests skill installation, the function checks the skill's feature flag before proceeding. If the flag is OFF, the install is rejected with an error.                                       |
+| Enforcement Point                  | Module                                                        | Effect                                                                                                                                                                                                      |
+| ---------------------------------- | ------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| **1. Client skill list**           | `resolveSkillStates()` in `config/skill-state.ts`             | Skills with flag OFF are excluded from the resolved list returned to clients (macOS skill list, settings UI). The skill never appears in the client.                                                        |
+| **2. System prompt skill catalog** | `appendSkillsCatalog()` in `prompts/system-prompt.ts`         | The model-visible `## Skills Catalog` section in the system prompt filters out flagged-off skills. The model cannot see or reference them.                                                                  |
+| **3. `skill_load` tool**           | `executeSkillLoad()` in `tools/skills/load.ts`                | If the model attempts to load a flagged-off skill by name, the tool returns an error: `"skill is currently unavailable (disabled by feature flag)"`.                                                        |
+| **4. Runtime tool projection**     | `projectSkillTools()` in `daemon/conversation-skill-tools.ts` | Even if a skill was previously active in a session (has `<loaded_skill>` markers in history), the per-turn projection drops it when the flag is OFF. Already-registered tools are unregistered.             |
+| **5. Included child skills**       | `executeSkillLoad()` in `tools/skills/load.ts`                | When a parent skill includes children via the `includes` directive, each child is independently checked against its feature flag. Flagged-off children are silently excluded from the loaded skill content. |
+| **6. Skill install gate**          | `installSkill()` in `daemon/handlers/skills.ts`               | When a client requests skill installation, the function checks the skill's feature flag before proceeding. If the flag is OFF, the install is rejected with an error.                                       |
 
 All six enforcement points derive the flag key via `skillFlagKey(skill)` — which returns `undefined` for ungated skills, short-circuiting the check — and then call `isAssistantFeatureFlagEnabled(flagKey, config)` for consistency.
 
@@ -655,7 +655,7 @@ All six enforcement points derive the flag key via `skillFlagKey(skill)` — whi
 | `src/config/skill-state.ts`                     | `skillFlagKey(skill)` — returns canonical flag key for skills with a `featureFlag` frontmatter field, `undefined` otherwise; `resolveSkillStates()` — enforcement point 1 |
 | `src/prompts/system-prompt.ts`                  | `appendSkillsCatalog()` — enforcement point 2                                                                                                                             |
 | `src/tools/skills/load.ts`                      | `executeSkillLoad()` — enforcement points 3 and 5                                                                                                                         |
-| `src/daemon/session-skill-tools.ts`             | `projectSkillTools()` — enforcement point 4                                                                                                                               |
+| `src/daemon/conversation-skill-tools.ts`        | `projectSkillTools()` — enforcement point 4                                                                                                                               |
 | `src/config/schema.ts`                          | `assistantFeatureFlagValues` field definition in `AssistantConfig` (Zod schema)                                                                                           |
 | `src/daemon/handlers/skills.ts`                 | `listSkills()` — uses `resolveSkillStates()` for client responses; `installSkill()` — enforcement point 6                                                                 |
 | `meta/feature-flags/feature-flag-registry.json` | Unified feature flag registry (repo root) — all declared flags with scope, label, default values, and descriptions                                                        |
@@ -688,7 +688,7 @@ graph LR
 
     subgraph "~/.vellum/workspace/data/db/assistant.db (SQLite + WAL)"
         direction TB
-        CONV["conversations<br/>───────────────<br/>id, title, timestamps<br/>token counts, estimated cost<br/>context_summary (compaction)<br/>thread_type: 'standard' | 'private'<br/>memory_scope_id: 'default' | 'private:&lt;uuid&gt;'"]
+        CONV["conversations<br/>───────────────<br/>id, title, timestamps<br/>token counts, estimated cost<br/>context_summary (compaction)<br/>conversation_type: 'standard' | 'private'<br/>memory_scope_id: 'default' | 'private:&lt;uuid&gt;'"]
         MSG["messages<br/>───────────────<br/>id, conversation_id (FK)<br/>role: user | assistant<br/>content: JSON array<br/>created_at"]
         TOOL["tool_invocations<br/>───────────────<br/>tool_name, input, result<br/>decision, risk_level<br/>duration_ms"]
         SEG["memory_segments<br/>───────────────<br/>Text chunks for retrieval<br/>Linked to messages<br/>token_estimate per segment"]
@@ -783,12 +783,12 @@ All client-server communication uses HTTP for request/response operations and Se
 
 The daemon emits two distinct error message types via SSE:
 
-| Message type    | Scope          | Purpose                                                                                                        | Payload                                                                       |
-| --------------- | -------------- | -------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
-| `session_error` | Session-scoped | Typed, actionable failures during chat/session runtime (e.g., provider network error, rate limit, API failure) | `sessionId`, `code` (typed enum), `userMessage`, `retryable`, `debugDetails?` |
-| `error`         | Global         | Generic, non-session failures (e.g., daemon startup errors, unknown message types)                             | `message` (string)                                                            |
+| Message type         | Scope          | Purpose                                                                                                        | Payload                                                                       |
+| -------------------- | -------------- | -------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
+| `conversation_error` | Session-scoped | Typed, actionable failures during chat/session runtime (e.g., provider network error, rate limit, API failure) | `sessionId`, `code` (typed enum), `userMessage`, `retryable`, `debugDetails?` |
+| `error`              | Global         | Generic, non-session failures (e.g., daemon startup errors, unknown message types)                             | `message` (string)                                                            |
 
-**Design rationale:** `session_error` carries structured metadata (error code, retryable flag, debug details) so the client can present actionable UI — a toast with retry/dismiss buttons — rather than a generic error banner. The older `error` type is retained for backward compatibility with non-session contexts.
+**Design rationale:** `conversation_error` carries structured metadata (error code, retryable flag, debug details) so the client can present actionable UI — a toast with retry/dismiss buttons — rather than a generic error banner. The older `error` type is retained for backward compatibility with non-session contexts.
 
 ### Session Error Codes
 
@@ -806,7 +806,7 @@ The daemon emits two distinct error message types via SSE:
 
 ### Error Classification
 
-The daemon classifies errors via `classifySessionError()` in `session-error.ts`. Before classification, `isUserCancellation()` checks whether the error is a user-initiated abort (active abort signal or `AbortError`); if so, the daemon emits `generation_cancelled` instead of `session_error` — cancel never surfaces a session-error toast.
+The daemon classifies errors via `classifyConversationError()` in `conversation-error.ts`. Before classification, `isUserCancellation()` checks whether the error is a user-initiated abort (active abort signal or `AbortError`); if so, the daemon emits `generation_cancelled` instead of `conversation_error` — cancel never surfaces a conversation-error toast.
 
 Classification uses a two-tier strategy:
 
@@ -819,45 +819,45 @@ Debug details are capped at 4,000 characters to prevent oversized payloads.
 
 ```mermaid
 sequenceDiagram
-    participant Daemon as Daemon (session-error.ts)
+    participant Daemon as Daemon (conversation-error.ts)
     participant DC as DaemonClient (Swift)
     participant VM as ChatViewModel
     participant UI as ChatView (toast)
 
     Note over Daemon: LLM call fails or<br/>processing error occurs
-    Daemon->>Daemon: classifySessionError(error, ctx)
-    Daemon->>DC: session_error {sessionId, code,<br/>userMessage, retryable, debugDetails?}
+    Daemon->>Daemon: classifyConversationError(error, ctx)
+    Daemon->>DC: conversation_error {sessionId, code,<br/>userMessage, retryable, debugDetails?}
     DC->>DC: broadcast to all subscribers
     DC->>VM: subscribe() stream delivers message
-    VM->>VM: set sessionError property<br/>clear isThinking / isCancelling
-    VM-->>UI: @Published sessionError observed
+    VM->>VM: set conversationError property<br/>clear isThinking / isCancelling
+    VM-->>UI: @Published conversationError observed
 
-    UI->>UI: show sessionErrorToast<br/>[Retry] [Dismiss] [Copy Debug Info?]
+    UI->>UI: show conversationErrorToast<br/>[Retry] [Dismiss] [Copy Debug Info?]
 
     alt User taps Retry (retryable == true)
-        UI->>VM: retryAfterSessionError()
-        VM->>VM: dismissSessionError()<br/>+ regenerateLastMessage()
+        UI->>VM: retryAfterConversationError()
+        VM->>VM: dismissConversationError()<br/>+ regenerateLastMessage()
         VM->>DC: regenerate {sessionId}
         DC->>Daemon: HTTP POST /v1/messages
     else User taps Dismiss
-        UI->>VM: dismissSessionError()
-        VM->>VM: clear sessionError + errorText
+        UI->>VM: dismissConversationError()
+        VM->>VM: clear conversationError + errorText
     end
 ```
 
-1. **Daemon** encounters a session-scoped failure, classifies it via `classifySessionError()`, and sends a `session_error` SSE event with the session ID, typed error code, user-facing message, retryable flag, and optional debug details. Session-scoped failures emit _only_ `session_error` (never the generic `error` type) to prevent cross-session bleed.
-2. **ChatViewModel** receives the error via DaemonClient's `subscribe()` stream (each view model gets an independent stream), sets the `sessionError` property, and transitions out of the streaming/loading state so the UI is interactive. If the error arrives during an active cancel (`wasCancelling == true`), it is suppressed — cancel only shows `generation_cancelled` behavior.
-3. **ChatView** observes the published `sessionError` and displays an actionable toast with a category-specific icon and accent color:
-   - **Retry** (shown when `retryable` is true): calls `retryAfterSessionError()`, which clears the error and sends a `regenerate` message to the daemon.
+1. **Daemon** encounters a session-scoped failure, classifies it via `classifyConversationError()`, and sends a `conversation_error` SSE event with the session ID, typed error code, user-facing message, retryable flag, and optional debug details. Session-scoped failures emit _only_ `conversation_error` (never the generic `error` type) to prevent cross-session bleed.
+2. **ChatViewModel** receives the error via DaemonClient's `subscribe()` stream (each view model gets an independent stream), sets the `conversationError` property, and transitions out of the streaming/loading state so the UI is interactive. If the error arrives during an active cancel (`wasCancelling == true`), it is suppressed — cancel only shows `generation_cancelled` behavior.
+3. **ChatView** observes the published `conversationError` and displays an actionable toast with a category-specific icon and accent color:
+   - **Retry** (shown when `retryable` is true): calls `retryAfterConversationError()`, which clears the error and sends a `regenerate` message to the daemon.
    - **Copy Debug Info** (shown when `debugDetails` is non-nil): copies structured debug information to the clipboard for bug reports.
-   - **Dismiss (X)**: calls `dismissSessionError()` to clear the error without retrying.
+   - **Dismiss (X)**: calls `dismissConversationError()` to clear the error without retrying.
 4. If the error is not retryable, the Retry button is hidden and the user can only dismiss.
 
 ---
 
 ## Context Overflow Recovery
 
-The session loop implements a deterministic overflow convergence pipeline that recovers from context-too-large provider rejections without surfacing errors to the user. Instead of the previous behavior where a `CONTEXT_TOO_LARGE` error was emitted as a `session_error`, the pipeline iteratively reduces the context payload until it fits within the provider's limit.
+The session loop implements a deterministic overflow convergence pipeline that recovers from context-too-large provider rejections without surfacing errors to the user. Instead of the previous behavior where a `CONTEXT_TOO_LARGE` error was emitted as a `conversation_error`, the pipeline iteratively reduces the context payload until it fits within the provider's limit.
 
 ### Two-Phase Architecture
 
@@ -882,15 +882,15 @@ After each tier, the reducer re-estimates tokens. If the estimate is within budg
 
 When all four reducer tiers are exhausted and the provider still rejects, the overflow policy resolver (`context-overflow-policy.ts`) determines the next action based on config and session interactivity:
 
-| Session Type    | Config Policy           | Action                                                                                                 |
-| --------------- | ----------------------- | ------------------------------------------------------------------------------------------------------ |
-| Interactive     | `"summarize"` (default) | `request_user_approval` — prompt the user via `PermissionPrompter` before compressing the latest turn  |
-| Non-interactive | `"truncate"` (default)  | `auto_compress_latest_turn` — compress without asking                                                  |
-| Any             | `"drop"`                | `fail_gracefully` — fall through to the final context-overflow fallback, which emits a `session_error` |
+| Session Type    | Config Policy           | Action                                                                                                      |
+| --------------- | ----------------------- | ----------------------------------------------------------------------------------------------------------- |
+| Interactive     | `"summarize"` (default) | `request_user_approval` — prompt the user via `PermissionPrompter` before compressing the latest turn       |
+| Non-interactive | `"truncate"` (default)  | `auto_compress_latest_turn` — compress without asking                                                       |
+| Any             | `"drop"`                | `fail_gracefully` — fall through to the final context-overflow fallback, which emits a `conversation_error` |
 
 **Approval gate:** For interactive sessions, the pipeline uses `requestCompressionApproval()` in `context-overflow-approval.ts`, which presents a confirmation prompt through the existing `PermissionPrompter` flow (`POST /v1/confirm`). The prompt uses a reserved pseudo tool name (`context_overflow_compression`) so the UI can display a meaningful label. The decision is one-shot per overflow (no "always allow" option).
 
-**Deny handling:** If the user declines compression, the session emits a graceful assistant explanation message ("The conversation has grown too long...") instead of a `session_error`. The deny message is persisted to conversation history and delivered via `assistant_text_delta` events, so the user sees a normal chat bubble rather than an error toast. The turn ends cleanly without triggering the error classification pipeline.
+**Deny handling:** If the user declines compression, the session emits a graceful assistant explanation message ("The conversation has grown too long...") instead of a `conversation_error`. The deny message is persisted to conversation history and delivered via `assistant_text_delta` events, so the user sees a normal chat bubble rather than an error toast. The turn ends cleanly without triggering the error classification pipeline.
 
 ### Config
 
@@ -911,14 +911,14 @@ All overflow recovery settings live under `contextWindow.overflowRecovery` in th
 | `src/daemon/context-overflow-reducer.ts`  | Tiered reducer: four-tier pipeline with idempotent steps and cumulative state    |
 | `src/daemon/context-overflow-policy.ts`   | Overflow policy resolver: maps config + interactivity to concrete action         |
 | `src/daemon/context-overflow-approval.ts` | Approval gate: prompts user for latest-turn compression via `PermissionPrompter` |
-| `src/daemon/session-agent-loop.ts`        | Integration: preflight budget check, convergence loop, approval/deny flow        |
+| `src/daemon/conversation-agent-loop.ts`   | Integration: preflight budget check, convergence loop, approval/deny flow        |
 | `src/config/core-schema.ts`               | `ContextOverflowRecoveryConfigSchema` with defaults and validation               |
 
 ---
 
 ## Task Routing — Voice Source Bypass and Escalation
 
-When a task is submitted via `task_submit`, the daemon classifies it to determine routing. Voice-sourced tasks and slash command candidates bypass the classifier entirely for lower latency and more predictable routing.
+When a task is submitted via `task_submit`, the daemon classifies it to determine routing. Voice-sourced tasks and built-in slash commands bypass the classifier entirely for lower latency and more predictable routing.
 
 ```mermaid
 graph TB
@@ -927,7 +927,7 @@ graph TB
     end
 
     subgraph "Routing Decision"
-        SLASH_CHECK{"Slash candidate?<br/>(parseSlashCandidate)"}
+        SLASH_CHECK{"Built-in slash command?<br/>(resolveSlash)"}
         VOICE_CHECK{"source === 'voice'?"}
         CLASSIFIER["Classifier<br/>Haiku-4.5 tool call<br/>+ heuristic fallback"]
         CU_ROUTE["Route: computer_use<br/>→ CU session"]
@@ -939,7 +939,7 @@ graph TB
     end
 
     SUBMIT --> SLASH_CHECK
-    SLASH_CHECK -->|"Yes (/skill-id)"| QA_ROUTE
+    SLASH_CHECK -->|"Yes (/model, /status, etc.)"| QA_ROUTE
     SLASH_CHECK -->|"No"| VOICE_CHECK
     VOICE_CHECK -->|"Yes"| QA_ROUTE
     VOICE_CHECK -->|"No"| CLASSIFIER
@@ -1006,29 +1006,25 @@ graph TB
 
 ## Slash Command Resolution
 
-When a user message enters the daemon (via `processMessage` or the queue drain path), it passes through slash command resolution before persistence or agent execution.
+When a user message enters the daemon (via `processMessage` or the queue drain path), it passes through `resolveSlash()` before persistence or agent execution. Resolution uses direct string matching against a fixed set of built-in commands.
 
 ```mermaid
 graph TB
     INPUT["User input"]
-    PARSE{"parseSlashCandidate"}
-    RESOLVE{"resolveSlashSkillCommand"}
-    NONE["Normal flow<br/>persist + agent loop"]
-    KNOWN["Rewrite to skill prompt<br/>persist + agent loop"]
-    UNKNOWN["Deterministic response<br/>list available commands<br/>no agent loop"]
-
-    INPUT --> PARSE
-    PARSE -->|"Not a slash candidate"| NONE
-    PARSE -->|"Valid candidate"| RESOLVE
-    RESOLVE -->|"Known skill ID"| KNOWN
-    RESOLVE -->|"Unknown ID"| UNKNOWN
+    RESOLVE{"resolveSlash()<br/>direct string matching"}
+    PASSTHROUGH["Normal flow<br/>persist + agent loop"]
+    HANDLED["Deterministic response<br/>assistant_text_delta + message_complete<br/>no agent loop"]
+
+    INPUT --> RESOLVE
+    RESOLVE -->|"kind: passthrough"| PASSTHROUGH
+    RESOLVE -->|"kind: unknown<br/>(/model, /status, /commands, /pair,<br/>/models, provider shortcuts)"| HANDLED
 ```
 
 Key behaviors:
 
-- **Known**: Content is rewritten via `rewriteKnownSlashCommandPrompt` to instruct the model to invoke the skill. Trailing arguments are preserved.
-- **Unknown**: A deterministic `assistant_text_delta` + `message_complete` is emitted listing available slash commands. No message persistence or model call occurs.
-- **Queue**: Queued messages receive the same slash resolution. Unknown slash commands in the queue emit their response and continue draining without stalling.
+- **Built-in commands**: `/model`, `/models`, `/status`, `/commands`, `/pair`, and provider shortcuts (`/opus`, `/sonnet`, `/gpt4`, etc.) are handled directly by `resolveSlash()`. A deterministic `assistant_text_delta` + `message_complete` is emitted. No message persistence or model call occurs.
+- **Passthrough**: Any input that does not match a built-in command passes through to the normal agent loop, including slash-like tokens that are not recognized.
+- **Queue**: Queued messages receive the same slash resolution.
 
 ---
 
@@ -1098,7 +1094,7 @@ graph TB
 - `evaluate_typescript_code` always forces `sandbox.enabled = true` regardless of global config.
 - Snippet contract: must export `default` or `run` with signature `(input: unknown) => unknown | Promise<unknown>`.
 - Managed-store writes are atomic (tmp file + rename) to prevent partial `SKILL.md` or `SKILLS.md` files.
-- After persist or delete, the file watcher triggers session eviction; the next turn runs in a fresh session. The model's system prompt instructs it to continue normally.
+- After persist or delete, the file watcher triggers conversation eviction; the next turn runs in a fresh conversation. The model's system prompt instructs it to continue normally.
 - macOS UI shows Inspect and Delete controls for managed skills only (source = "managed").
 - `skill_load` validates the recursive include graph (via `include-graph.ts`) before emitting output. Missing children and cycles produce `isError: true` with no `<loaded_skill>` marker. Valid includes produce an "Included Skills (immediate)" metadata section showing child ID, name, description, and path.
 
@@ -1162,7 +1158,7 @@ Rules enforced by guard tests:
 - Direct gateway `curl` + manual bearer headers are for control-plane writes/actions, not retrieval reads.
 - Bundled skill docs must not instruct direct keychain lookups (`security find-generic-password`, `secret-tool`) for retrieval.
 - `host_bash` is not used for Vellum CLI retrieval commands unless intentionally allowlisted.
-- Outbound credentialed API calls prefer proxied execution (`bash` with `network_mode: "proxied"` + `credential_ids`) so credentials are injected by policy-aware plumbing instead of copied into commands.
+- Outbound credentialed API calls use CES tools (`make_authenticated_request`, `run_authenticated_command`) so credential materialization happens in a separate process. Command output (stdout/stderr) is forwarded back to the assistant and may contain credential values if the command echoes them, so the isolation covers injection, not output. `host_bash` is available as a user-approved escape hatch but is outside the strong secrecy guarantee.
 
 ### Skill Directory Structure
 
@@ -1196,12 +1192,11 @@ The following capabilities ship as bundled skills in `assistant/src/config/bundl
 ```mermaid
 graph TB
     subgraph "Activation Sources"
-        SLASH["Slash command<br/>/skill-id → preactivate"]
         MARKER["&lt;loaded_skill id=&quot;...&quot; /&gt;<br/>marker in conversation history"]
         CONFIG["Config / session<br/>preactivatedSkillIds"]
     end
 
-    subgraph "Per-Turn Projection (session-skill-tools.ts)"
+    subgraph "Per-Turn Projection (conversation-skill-tools.ts)"
         DERIVE["deriveActiveSkills(history)<br/>scan all messages for markers"]
         UNION["Union: context-derived ∪ preactivated"]
         DIFF["Diff vs previous turn"]
@@ -1218,7 +1213,6 @@ graph TB
         PROVIDER["LLM Provider<br/>receives full tool list"]
     end
 
-    SLASH --> CONFIG
     MARKER --> DERIVE
     CONFIG --> UNION
     DERIVE --> UNION
@@ -1233,7 +1227,7 @@ graph TB
     RESOLVE --> PROVIDER
 ```
 
-**Internal preactivation**: Some bundled skills are preactivated programmatically rather than by user slash commands or model discovery. For example, desktop sessions set `preactivatedSkillIds: ['computer-use']`, causing `projectSkillTools()` to load the 11 `computer_use_*` tool definitions from the bundled skill's `TOOLS.json` on the first turn. These proxy tools forward actions to the connected macOS client via `HostCuProxy`.
+**Internal preactivation**: Some bundled skills are preactivated programmatically rather than by model discovery. For example, desktop sessions set `preactivatedSkillIds: ['computer-use']`, causing `projectSkillTools()` to load the 11 `computer_use_*` tool definitions from the bundled skill's `TOOLS.json` on the first turn. These proxy tools forward actions to the connected macOS client via `HostCuProxy`.
 
 ### Skill Tool Execution
 
@@ -1276,7 +1270,7 @@ graph TB
 | `assistant/src/skills/tool-manifest.ts`             | `TOOLS.json` parser and validator                                                          |
 | `assistant/src/skills/active-skill-tools.ts`        | `deriveActiveSkills()` — scans history for `<loaded_skill>` markers                        |
 | `assistant/src/skills/include-graph.ts`             | Include graph builder: `indexCatalogById()`, `validateIncludes()`, cycle/missing detection |
-| `assistant/src/daemon/session-skill-tools.ts`       | `projectSkillTools()` — per-turn projection, register/unregister lifecycle                 |
+| `assistant/src/daemon/conversation-skill-tools.ts`  | `projectSkillTools()` — per-turn projection, register/unregister lifecycle                 |
 | `assistant/src/tools/skills/skill-tool-factory.ts`  | `createSkillToolsFromManifest()` — manifest entries to Tool objects                        |
 | `assistant/src/tools/skills/skill-script-runner.ts` | Host runner: dynamic import + `run()` call                                                 |
 | `assistant/src/tools/skills/sandbox-runner.ts`      | Sandbox runner: isolated subprocess execution                                              |
@@ -1480,7 +1474,7 @@ The `tool_permission_simulate` HTTP endpoint lets clients dry-run a tool invocat
 - The daemon runs `classifyRisk()` and `check()` against the live trust rules, then returns the decision (`allow`, `deny`, or `prompt`), risk level, reason, matched rule ID, and (when decision is `prompt`) the full `promptPayload` with allowlist/scope options.
 - **Simulation-only allow/deny**: A simulated `allow` or `deny` decision does not persist any state. No trust rules are created or modified.
 - **Always-allow persistence**: When the tester UI's "Always Allow" action is used, the client sends a separate `add_trust_rule` message that persists the rule to `trust.json`, identical to the existing confirmation flow.
-- **Private-thread override**: When `forcePromptSideEffects` is true, side-effect tools that would normally be auto-allowed are promoted to `prompt`.
+- **Private-conversation override**: When `forcePromptSideEffects` is true, side-effect tools that would normally be auto-allowed are promoted to `prompt`.
 - **Non-interactive override**: When `isInteractive` is false, `prompt` decisions are converted to `deny` (no client available to approve).
 
 ---
@@ -1693,7 +1687,7 @@ The assistant-events system provides a single, shared publish path that fans out
 graph TB
     subgraph "Event Sources"
         direction TB
-        SESSION["Session process<br/>(session-process.ts)"]
+        SESSION["Session process<br/>(conversation-process.ts)"]
         HTTP_RUN["HTTP Run path<br/>(run-orchestrator.ts)"]
     end
 
@@ -1725,13 +1719,13 @@ graph TB
 
 Every event published through the hub is wrapped in an `AssistantEvent` (defined in `runtime/assistant-event.ts`):
 
-| Field         | Type                | Description                                           |
-| ------------- | ------------------- | ----------------------------------------------------- |
-| `id`          | `string` (UUID)     | Globally unique event identifier                      |
-| `assistantId` | `string`            | Logical assistant identifier (`"self"` for HTTP runs) |
-| `sessionId`   | `string?`           | Resolved conversation ID when available               |
-| `emittedAt`   | `string` (ISO-8601) | Server-side timestamp                                 |
-| `message`     | `ServerMessage`     | The outbound message payload                          |
+| Field            | Type                | Description                                           |
+| ---------------- | ------------------- | ----------------------------------------------------- |
+| `id`             | `string` (UUID)     | Globally unique event identifier                      |
+| `assistantId`    | `string`            | Logical assistant identifier (`"self"` for HTTP runs) |
+| `conversationId` | `string?`           | Resolved conversation ID when available               |
+| `emittedAt`      | `string` (ISO-8601) | Server-side timestamp                                 |
+| `message`        | `ServerMessage`     | The outbound message payload                          |
 
 ### SSE Frame Format
 
@@ -1777,8 +1771,8 @@ The notification module (`assistant/src/notifications/`) uses a signal-based arc
 ```
 Producer → NotificationSignal → Candidate Generation → Decision Engine (LLM) → Deterministic Checks → Broadcaster → Conversation Pairing → Adapters → Delivery
                                                               ↑                                                            ↓
-                                                      Preference Summary                                    notification_thread_created SSE event
-                                                      Thread Candidates                                     (creation-only — not emitted on reuse)
+                                                      Preference Summary                                    notification_conversation_created SSE event
+                                                      Conversation Candidates                                (creation-only — not emitted on reuse)
 ```
 
 ### Channel Policy Registry
@@ -1787,24 +1781,24 @@ Producer → NotificationSignal → Candidate Generation → Decision Engine (LL
 
 - **`deliveryEnabled`** — whether the channel can receive notification deliveries. The `NotificationChannel` type is derived from this flag: only channels with `deliveryEnabled: true` are valid notification targets.
 - **`conversationStrategy`** — how the notification pipeline materializes conversations for the channel:
-  - `start_new_conversation` — creates a fresh conversation per delivery (e.g. vellum desktop/mobile threads)
+  - `start_new_conversation` — creates a fresh conversation per delivery (e.g. vellum desktop/mobile conversations)
   - `continue_existing_conversation` — intended to append to an existing channel-scoped conversation; currently materializes a background audit conversation per delivery (e.g. Telegram)
   - `not_deliverable` — channel cannot receive notifications (e.g. phone)
 
 Helper functions: `getDeliverableChannels()`, `getChannelPolicy()`, `isNotificationDeliverable()`, `getConversationStrategy()`.
 
-### Conversation Pairing and Thread Routing
+### Conversation Pairing and Conversation Routing
 
-Every notification delivery materializes a conversation + seed message **before** the adapter sends it (`conversation-pairing.ts`). The pairing function now accepts a `threadAction` from the decision engine:
+Every notification delivery materializes a conversation + seed message **before** the adapter sends it (`conversation-pairing.ts`). The pairing function now accepts a `conversationAction` from the decision engine:
 
-- **`reuse_existing`**: Looks up the target conversation. If valid (exists with `source: 'notification'`), the seed message is appended to the existing thread. If invalid, falls back to creating a new conversation with `threadDecisionFallbackUsed: true`.
+- **`reuse_existing`**: Looks up the target conversation. If valid (exists with `source: 'notification'`), the seed message is appended to the existing conversation. If invalid, falls back to creating a new conversation with `conversationDecisionFallbackUsed: true`.
 - **`start_new` (default)**: Creates a fresh conversation per delivery.
 
 This ensures:
 
 1. Every delivery has an auditable conversation trail in the conversations table
-2. The macOS/iOS client can deep-link directly into the notification thread
-3. Delivery audit rows in `notification_deliveries` carry `conversation_id`, `message_id`, `conversation_strategy`, `thread_action`, `thread_target_conversation_id`, and `thread_decision_fallback_used` columns
+2. The macOS/iOS client can deep-link directly into the notification conversation
+3. Delivery audit rows in `notification_deliveries` carry `conversation_id`, `message_id`, `conversation_strategy`, `conversation_action`, `conversation_target_id`, and `conversation_fallback_used` columns
 
 The pairing function (`pairDeliveryWithConversation`) is resilient — errors are caught and logged without breaking the delivery pipeline.
 
@@ -1812,53 +1806,53 @@ The pairing function (`pairDeliveryWithConversation`) is resilient — errors ar
 
 The notification pipeline uses a single conversation materialization path across producers:
 
-1. **Canonical pipeline** (`emitNotificationSignal` → decision engine → broadcaster → conversation pairing → adapters): The broadcaster pairs each delivery with a conversation, then dispatches a `notification_intent` SSE event via the Vellum adapter. The payload includes `deepLinkMetadata` (e.g. `{ conversationId, messageId }`) so the macOS/iOS client can deep-link to the relevant context when the user taps the notification. When `messageId` is present, the client scrolls to that specific message within the thread (message-level anchoring).
-2. **Guardian bookkeeping** (`dispatchGuardianQuestion`): Guardian dispatch creates `guardian_action_request` / `guardian_action_delivery` audit rows derived from pipeline delivery results and the per-dispatch `onThreadCreated` callback — there is no separate thread-creation path.
+1. **Canonical pipeline** (`emitNotificationSignal` → decision engine → broadcaster → conversation pairing → adapters): The broadcaster pairs each delivery with a conversation, then dispatches a `notification_intent` SSE event via the Vellum adapter. The payload includes `deepLinkMetadata` (e.g. `{ conversationId, messageId }`) so the macOS/iOS client can deep-link to the relevant context when the user taps the notification. When `messageId` is present, the client scrolls to that specific message within the conversation (message-level anchoring).
+2. **Guardian bookkeeping** (`dispatchGuardianQuestion`): Guardian dispatch creates `guardian_action_request` / `guardian_action_delivery` audit rows derived from pipeline delivery results and the per-dispatch `onConversationCreated` callback — there is no separate conversation-creation path.
 
-### Thread Surfacing via `notification_thread_created` (Creation-Only)
+### Conversation Surfacing via `notification_conversation_created` (Creation-Only)
 
-The `notification_thread_created` SSE event is emitted **only when a brand-new conversation is created** by the broadcaster. Reusing an existing thread does not trigger this event — the macOS/iOS client already knows about the conversation from the original creation. This is enforced in `broadcaster.ts` by gating on `pairing.createdNewConversation === true`.
+The `notification_conversation_created` SSE event is emitted **only when a brand-new conversation is created** by the broadcaster. Reusing an existing conversation does not trigger this event — the macOS/iOS client already knows about the conversation from the original creation. This is enforced in `broadcaster.ts` by gating on `pairing.createdNewConversation === true`.
 
-When a new vellum notification thread is created (strategy `start_new_conversation`), the broadcaster emits the event **immediately** (before waiting for slower channel deliveries like Telegram). This pushes the thread to the macOS/iOS client so it can display the notification thread in the sidebar and deep-link to it.
+When a new vellum notification conversation is created (strategy `start_new_conversation`), the broadcaster emits the event **immediately** (before waiting for slower channel deliveries like Telegram). This pushes the conversation to the macOS/iOS client so it can display the notification conversation in the sidebar and deep-link to it.
 
-### Thread-Created Events
+### Conversation-Created Events
 
-Two SSE push events surface new threads in the macOS/iOS client sidebar:
+Two SSE push events surface new conversations in the macOS/iOS client sidebar:
 
-- **`notification_thread_created`** — Emitted by `broadcaster.ts` when a notification delivery **creates** a new vellum conversation (strategy `start_new_conversation`, `createdNewConversation: true`). **Not** emitted when a thread is reused. Payload: `{ conversationId, title, sourceEventName }`.
-- **`task_run_thread_created`** — Emitted by `work-item-runner.ts` when a task run creates a conversation. Payload: `{ conversationId, workItemId, title }`.
+- **`notification_conversation_created`** — Emitted by `broadcaster.ts` when a notification delivery **creates** a new vellum conversation (strategy `start_new_conversation`, `createdNewConversation: true`). **Not** emitted when a conversation is reused. Payload: `{ conversationId, title, sourceEventName }`.
+- **`task_run_conversation_created`** — Emitted by `work-item-runner.ts` when a task run creates a conversation. Payload: `{ conversationId, workItemId, title }`.
 
-All events follow the same pattern: the daemon creates a server-side conversation, persists an initial message, and broadcasts the SSE event so the macOS `ThreadManager` can create a visible thread in the sidebar.
+All events follow the same pattern: the daemon creates a server-side conversation, persists an initial message, and broadcasts the SSE event so the macOS `ConversationManager` can create a visible conversation in the sidebar.
 
-### Thread Routing Decision Flow
+### Conversation Routing Decision Flow
 
-The decision engine produces per-channel thread actions using a candidate-driven approach:
+The decision engine produces per-channel conversation actions using a candidate-driven approach:
 
-1. **Candidate generation** (`thread-candidates.ts`): Queries recent notification-sourced conversations (24-hour window, up to 5 per channel) and enriches them with guardian context (pending request counts).
+1. **Candidate generation** (`conversation-candidates.ts`): Queries recent notification-sourced conversations (24-hour window, up to 5 per channel) and enriches them with guardian context (pending request counts).
 2. **LLM decision**: The candidate set is serialized into the system prompt. The LLM chooses `start_new` or `reuse_existing` (with a candidate `conversationId`) per channel.
-3. **Strict validation** (`validateThreadActions`): Reuse targets must exist in the candidate set. Invalid targets are downgraded to `start_new`.
-4. **Pairing execution**: `pairDeliveryWithConversation` executes the thread action — appending to an existing conversation on reuse, creating a new one otherwise.
-5. **Creation-only gating**: `notification_thread_created` fires only on actual creation, not on reuse.
-6. **Audit trail**: Thread actions are persisted in both `notification_decisions.validation_results` and `notification_deliveries` columns (`thread_action`, `thread_target_conversation_id`, `thread_decision_fallback_used`).
+3. **Strict validation** (`validateConversationActions`): Reuse targets must exist in the candidate set. Invalid targets are downgraded to `start_new`.
+4. **Pairing execution**: `pairDeliveryWithConversation` executes the conversation action — appending to an existing conversation on reuse, creating a new one otherwise.
+5. **Creation-only gating**: `notification_conversation_created` fires only on actual creation, not on reuse.
+6. **Audit trail**: Conversation actions are persisted in both `notification_decisions.validation_results` and `notification_deliveries` columns (`conversation_action`, `conversation_target_id`, `conversation_fallback_used`).
 
-### Guardian Call Thread Affinity
+### Guardian Call Conversation Affinity
 
-When a guardian question originates from an active phone call (`callSessionId` present on the signal), the decision engine enforces thread affinity so all questions within the same call land in one vellum thread:
+When a guardian question originates from an active phone call (`callSessionId` present on the signal), the decision engine enforces conversation affinity so all questions within the same call land in one vellum conversation:
 
-- **First question in a call** (no `conversationAffinityHint`): `enforceGuardianCallThreadAffinity` forces `start_new` for the vellum channel, creating a dedicated thread for the call.
+- **First question in a call** (no `conversationAffinityHint`): `enforceGuardianCallConversationAffinity` forces `start_new` for the vellum channel, creating a dedicated conversation for the call.
 - **Subsequent questions in the same call** (affinity hint already set by `dispatchGuardianQuestion`): The guard is a no-op, and `enforceConversationAffinity` routes to `reuse_existing` using the hint's `conversationId`.
 
-This guard runs **before** `enforceConversationAffinity` in the post-decision chain so the two cooperate: the first dispatch creates the thread, and subsequent dispatches reuse it via the affinity hint that `dispatchGuardianQuestion` sets after observing the first delivery's `conversationId`.
+This guard runs **before** `enforceConversationAffinity` in the post-decision chain so the two cooperate: the first dispatch creates the conversation, and subsequent dispatches reuse it via the affinity hint that `dispatchGuardianQuestion` sets after observing the first delivery's `conversationId`.
 
-### Guardian Multi-Request Disambiguation in Reused Threads
+### Guardian Multi-Request Disambiguation in Reused Conversations
 
-When the decision engine routes multiple guardian questions to the same conversation (via `reuse_existing`), those questions share a single thread. The guardian disambiguates which question they are answering using **request-code prefixes**:
+When the decision engine routes multiple guardian questions to the same conversation (via `reuse_existing`), those questions share a single conversation. The guardian disambiguates which question they are answering using **request-code prefixes**:
 
 - **Single pending delivery**: Matched automatically (single-match fast path).
 - **Multiple pending deliveries**: The guardian must prefix their reply with the 6-char hex request code (e.g. `A1B2C3 yes, allow it`). Case-insensitive matching.
 - **No match**: A disambiguation message is sent listing all active request codes.
 
-This invariant is enforced identically on mac/vellum (`session-process.ts`) and Telegram (`inbound-message-handler.ts`). All disambiguation messages are generated through the guardian action message composer (LLM with deterministic fallback).
+This invariant is enforced identically on mac/vellum (`conversation-process.ts`) and Telegram (`inbound-message-handler.ts`). All disambiguation messages are generated through the guardian action message composer (LLM with deterministic fallback).
 
 ### Reminder Routing Metadata
 
@@ -1868,7 +1862,7 @@ Reminders carry optional `routingIntent` (`single_channel` | `multi_channel` | `
 
 Notifications are delivered to three channel types:
 
-- **Vellum (always connected)**: SSE via the daemon's broadcast mechanism. The `VellumAdapter` emits a `notification_intent` message with rendered copy and optional `deepLinkMetadata` (includes `conversationId` for thread navigation and `messageId` for message-level scroll anchoring).
+- **Vellum (always connected)**: SSE via the daemon's broadcast mechanism. The `VellumAdapter` emits a `notification_intent` message with rendered copy and optional `deepLinkMetadata` (includes `conversationId` for conversation navigation and `messageId` for message-level scroll anchoring).
 - **Telegram (when guardian binding exists)**: HTTP POST to the gateway's `/deliver/telegram` endpoint. Requires an active guardian binding for the assistant.
 
 Connected channels are resolved at signal emission time: vellum is always included, and binding-based channels (Telegram) are included only when an active guardian binding exists for the assistant.
@@ -1881,9 +1875,9 @@ Connected channels are resolved at signal emission time: vellum is always includ
 | `assistant/src/notifications/emit-signal.ts`                               | Single entry point for all producers; orchestrates the full pipeline                                                                  |
 | `assistant/src/notifications/decision-engine.ts`                           | LLM-based routing decisions with deterministic fallback                                                                               |
 | `assistant/src/notifications/deterministic-checks.ts`                      | Hard invariant checks (dedupe, source-active suppression, channel availability)                                                       |
-| `assistant/src/notifications/broadcaster.ts`                               | Dispatches decisions to channel adapters; emits `notification_thread_created` SSE event (creation-only)                               |
-| `assistant/src/notifications/conversation-pairing.ts`                      | Materializes conversation + message per delivery; executes thread reuse decisions                                                     |
-| `assistant/src/notifications/thread-candidates.ts`                         | Builds per-channel candidate set of recent conversations for the decision engine                                                      |
+| `assistant/src/notifications/broadcaster.ts`                               | Dispatches decisions to channel adapters; emits `notification_conversation_created` SSE event (creation-only)                         |
+| `assistant/src/notifications/conversation-pairing.ts`                      | Materializes conversation + message per delivery; executes conversation reuse decisions                                               |
+| `assistant/src/notifications/conversation-candidates.ts`                   | Builds per-channel candidate set of recent conversations for the decision engine                                                      |
 | `assistant/src/notifications/adapters/macos.ts`                            | Vellum adapter — broadcasts `notification_intent` via SSE with deep-link metadata                                                     |
 | `assistant/src/notifications/adapters/telegram.ts`                         | Telegram adapter — POSTs to gateway `/deliver/telegram`                                                                               |
 | `assistant/src/notifications/destination-resolver.ts`                      | Resolves per-channel endpoints (vellum SSE, Telegram chat ID from guardian binding)                                                   |
@@ -1893,7 +1887,7 @@ Connected channels are resolved at signal emission time: vellum is always includ
 | `assistant/src/config/bundled-skills/messaging/tools/send-notification.ts` | Explicit producer tool for user-requested notifications; emits signals into the same routing pipeline                                 |
 | `assistant/src/calls/guardian-dispatch.ts`                                 | Guardian question dispatch that reuses canonical notification pairing and records guardian delivery bookkeeping from pipeline results |
 
-**Audit trail (SQLite):** `notification_events` → `notification_decisions` (with `threadActions` in validation results) → `notification_deliveries` (with `conversation_id`, `message_id`, `conversation_strategy`, `thread_action`, `thread_target_conversation_id`, `thread_decision_fallback_used`)
+**Audit trail (SQLite):** `notification_events` → `notification_decisions` (with `conversationActions` in validation results) → `notification_deliveries` (with `conversation_id`, `message_id`, `conversation_strategy`, `conversation_action`, `conversation_target_id`, `conversation_fallback_used`)
 
 **Configuration:** `notifications.decisionModelIntent` in `config.json`.
 
@@ -1901,44 +1895,44 @@ Connected channels are resolved at signal emission time: vellum is always includ
 
 ## Storage Summary
 
-| What                                         | Where                                                             | Format                              | ORM/Driver                         | Retention                                                  |
-| -------------------------------------------- | ----------------------------------------------------------------- | ----------------------------------- | ---------------------------------- | ---------------------------------------------------------- |
-| API key                                      | macOS Keychain                                                    | Encrypted binary                    | `/usr/bin/security` CLI            | Permanent                                                  |
-| Credential secrets                           | macOS Keychain (or encrypted file fallback)                       | Encrypted binary                    | `secure-keys.ts` wrapper           | Permanent (until deleted via tool)                         |
-| Credential metadata                          | `~/.vellum/workspace/data/credentials/metadata.json`              | JSON                                | Atomic file write                  | Permanent (until deleted via tool)                         |
-| Integration OAuth tokens                     | macOS Keychain (or encrypted file fallback, via `secure-keys.ts`) | Encrypted binary                    | `TokenManager` auto-refresh        | Until disconnected or revoked                              |
-| User preferences                             | UserDefaults                                                      | plist                               | Foundation                         | Permanent                                                  |
-| Session logs                                 | `~/Library/.../logs/session-*.json`                               | JSON per session                    | Swift Codable                      | Unbounded                                                  |
-| Conversations & messages                     | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite + WAL                        | Drizzle ORM (Bun)                  | Permanent                                                  |
-| Memory segments                              | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent                                                  |
-| Extracted facts                              | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent, deduped                                         |
-| Embeddings                                   | `~/.vellum/workspace/data/db/assistant.db`                        | JSON float arrays                   | Drizzle ORM                        | Permanent                                                  |
-| Async job queue                              | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Completed jobs persist                                     |
-| Attachments                                  | `~/.vellum/workspace/data/db/assistant.db`                        | Base64 in SQLite                    | Drizzle ORM                        | Permanent                                                  |
-| Sandbox filesystem                           | `~/.vellum/workspace`                                             | Real filesystem tree                | Node FS APIs                       | Persistent across sessions                                 |
-| Tool permission rules                        | `~/.vellum/protected/trust.json`                                  | JSON                                | File I/O                           | Permanent                                                  |
-| Web users & assistants                       | PostgreSQL                                                        | Relational                          | Drizzle ORM (pg)                   | Permanent                                                  |
-| Trace events                                 | In-memory (TraceStore)                                            | Structured events                   | Swift ObservableObject             | Max 5,000 per session, ephemeral                           |
-| Media embed settings                         | `~/.vellum/workspace/config.json` (`ui.mediaEmbeds`)              | JSON                                | `WorkspaceConfigIO` (atomic merge) | Permanent                                                  |
-| Media embed MIME cache                       | In-memory (`ImageMIMEProbe`)                                      | `NSCache` (500 entries)             | HTTP HEAD                          | Ephemeral; cleared on app restart                          |
-| Tasks & task runs                            | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent                                                  |
-| Work items (Task Queue)                      | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; archived items retained                         |
-| Recurrence schedules & runs                  | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; supports cron and RRULE syntax                  |
-| Watchers & events                            | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent, cascade on watcher delete                       |
-| Proxy CA cert + key                          | `{dataDir}/proxy-ca/`                                             | PEM files (ca.pem, ca-key.pem)      | openssl CLI                        | Permanent (10-year validity)                               |
-| Proxy leaf certs                             | `{dataDir}/proxy-ca/issued/`                                      | PEM files per hostname              | openssl CLI, cached                | 1-year validity, re-issued on CA change                    |
-| Proxy sessions                               | In-memory (SessionManager)                                        | Map<ProxySessionId, ManagedSession> | Manual lifecycle                   | Ephemeral; 5min idle timeout, cleared on shutdown          |
-| Call sessions, events, pending questions     | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent, cascade on session delete                       |
-| Active call controllers                      | In-memory (CallState)                                             | Map<callSessionId, CallController>  | Manual lifecycle                   | Ephemeral; cleared on call end or destroy                  |
-| Guardian bindings                            | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; revoked bindings retained                       |
-| Channel verification sessions                | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; consumed/expired sessions retained              |
-| Guardian approval requests                   | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; decision outcome retained                       |
-| Contact invites                              | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; token hash stored, raw token never persisted    |
-| Contacts & channels                          | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; revoked/blocked contacts retained               |
-| Notification events                          | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; deduplicated by dedupeKey                       |
-| Notification decisions                       | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; FK to notification_events                       |
-| Notification deliveries                      | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; FK to notification_decisions                    |
-| Notification preferences                     | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; per-assistant conversational preferences        |
+| What                                     | Where                                                             | Format                              | ORM/Driver                         | Retention                                               |
+| ---------------------------------------- | ----------------------------------------------------------------- | ----------------------------------- | ---------------------------------- | ------------------------------------------------------- |
+| API key                                  | macOS Keychain                                                    | Encrypted binary                    | `/usr/bin/security` CLI            | Permanent                                               |
+| Credential secrets                       | macOS Keychain (or encrypted file fallback)                       | Encrypted binary                    | `secure-keys.ts` wrapper           | Permanent (until deleted via tool)                      |
+| Credential metadata                      | `~/.vellum/workspace/data/credentials/metadata.json`              | JSON                                | Atomic file write                  | Permanent (until deleted via tool)                      |
+| Integration OAuth tokens                 | macOS Keychain (or encrypted file fallback, via `secure-keys.ts`) | Encrypted binary                    | `TokenManager` auto-refresh        | Until disconnected or revoked                           |
+| User preferences                         | UserDefaults                                                      | plist                               | Foundation                         | Permanent                                               |
+| Session logs                             | `~/Library/.../logs/session-*.json`                               | JSON per session                    | Swift Codable                      | Unbounded                                               |
+| Conversations & messages                 | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite + WAL                        | Drizzle ORM (Bun)                  | Permanent                                               |
+| Memory segments                          | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent                                               |
+| Extracted facts                          | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent, deduped                                      |
+| Embeddings                               | `~/.vellum/workspace/data/db/assistant.db`                        | JSON float arrays                   | Drizzle ORM                        | Permanent                                               |
+| Async job queue                          | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Completed jobs persist                                  |
+| Attachments                              | `~/.vellum/workspace/data/db/assistant.db`                        | Base64 in SQLite                    | Drizzle ORM                        | Permanent                                               |
+| Sandbox filesystem                       | `~/.vellum/workspace`                                             | Real filesystem tree                | Node FS APIs                       | Persistent across sessions                              |
+| Tool permission rules                    | `~/.vellum/protected/trust.json`                                  | JSON                                | File I/O                           | Permanent                                               |
+| Web users & assistants                   | PostgreSQL                                                        | Relational                          | Drizzle ORM (pg)                   | Permanent                                               |
+| Trace events                             | In-memory (TraceStore)                                            | Structured events                   | Swift ObservableObject             | Max 5,000 per session, ephemeral                        |
+| Media embed settings                     | `~/.vellum/workspace/config.json` (`ui.mediaEmbeds`)              | JSON                                | `WorkspaceConfigIO` (atomic merge) | Permanent                                               |
+| Media embed MIME cache                   | In-memory (`ImageMIMEProbe`)                                      | `NSCache` (500 entries)             | HTTP HEAD                          | Ephemeral; cleared on app restart                       |
+| Tasks & task runs                        | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent                                               |
+| Work items (Task Queue)                  | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; archived items retained                      |
+| Recurrence schedules & runs              | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; supports cron and RRULE syntax               |
+| Watchers & events                        | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent, cascade on watcher delete                    |
+| Proxy CA cert + key                      | `{dataDir}/proxy-ca/`                                             | PEM files (ca.pem, ca-key.pem)      | openssl CLI                        | Permanent (10-year validity)                            |
+| Proxy leaf certs                         | `{dataDir}/proxy-ca/issued/`                                      | PEM files per hostname              | openssl CLI, cached                | 1-year validity, re-issued on CA change                 |
+| Proxy sessions                           | In-memory (SessionManager)                                        | Map<ProxySessionId, ManagedSession> | Manual lifecycle                   | Ephemeral; 5min idle timeout, cleared on shutdown       |
+| Call sessions, events, pending questions | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent, cascade on session delete                    |
+| Active call controllers                  | In-memory (CallState)                                             | Map<callSessionId, CallController>  | Manual lifecycle                   | Ephemeral; cleared on call end or destroy               |
+| Guardian bindings                        | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; revoked bindings retained                    |
+| Channel verification sessions            | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; consumed/expired sessions retained           |
+| Guardian approval requests               | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; decision outcome retained                    |
+| Contact invites                          | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; token hash stored, raw token never persisted |
+| Contacts & channels                      | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; revoked/blocked contacts retained            |
+| Notification events                      | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; deduplicated by dedupeKey                    |
+| Notification decisions                   | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; FK to notification_events                    |
+| Notification deliveries                  | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; FK to notification_decisions                 |
+| Notification preferences                 | `~/.vellum/workspace/data/db/assistant.db`                        | SQLite                              | Drizzle ORM                        | Permanent; per-assistant conversational preferences     |
 
 ### Sensitive Tool Output Placeholder Substitution
 
@@ -1973,7 +1967,7 @@ The daemon uses a single fixed internal scope constant — `DAEMON_INTERNAL_ASSI
 
 The guardian trust system uses a three-valued `TrustClass` — `'guardian'`, `'trusted_contact'`, or `'unknown'` — as the single vocabulary for actor trust classification across all channels and runtime paths. There is no legacy `actorRole` concept; all trust decisions flow through `TrustClass`.
 
-**`TrustContext`** (in `src/daemon/session-runtime-assembly.ts`) is the single runtime carrier for trust state on channel-originated turns. It carries `trustClass`, guardian identity fields, and requester metadata. The `guardianPrincipalId` field is typed as `?: string` (optional but non-nullable) — a principal ID is present when a guardian binding exists but is never `null`.
+**`TrustContext`** (in `src/daemon/conversation-runtime-assembly.ts`) is the single runtime carrier for trust state on channel-originated turns. It carries `trustClass`, guardian identity fields, and requester metadata. The `guardianPrincipalId` field is typed as `?: string` (optional but non-nullable) — a principal ID is present when a guardian binding exists but is never `null`.
 
 **Explicit trust gates:** `trustClass` is a **required** field in `ToolContext` (in `src/tools/types.ts`). Every tool execution must carry a trust classification — the field is not optional. This ensures trust-gated tool policies (guardian control-plane restrictions, host-tool blocking for untrusted actors) cannot be bypassed by omitting the classification.
 
@@ -1987,7 +1981,7 @@ The guardian trust system uses a three-valued `TrustClass` — `'guardian'`, `'t
 
 | File                                          | Purpose                                               |
 | --------------------------------------------- | ----------------------------------------------------- |
-| `src/daemon/session-runtime-assembly.ts`      | `TrustContext` type definition                        |
+| `src/daemon/conversation-runtime-assembly.ts` | `TrustContext` type definition                        |
 | `src/tools/types.ts`                          | `ToolContext.trustClass` (required trust gate)        |
 | `src/runtime/channel-retry-sweep.ts`          | Strict `trustClass` parser for retry sweep            |
 | `src/memory/channel-verification-sessions.ts` | `GuardianBinding` with required `guardianPrincipalId` |