@terreno/mcp 57.10.1 → 57.11.0

package/src/docs/versioned/next/reference/ai-harness.md

@@ -141,7 +141,7 @@ The `task` view passed to `run`: `{id, name, version, input, state, phase, attem
 | --- | --- |
 | `Harness.open({registry, runner?, models?, env?, priceMap?, streaming?, testHooks?})` | Checks requirements, rejects a duplicate `name@version` or agent name, ensures collections and indexes exist. `runner` defaults to `new InProcessRunner()`. See [open options](#open-options). |
 | `start()` | Throws, claiming nothing, when any non-terminal task uses a `name@version` missing from the registry (see [Versioning](#versioning)). Then starts the runner. Throws when already started. The runner recovers expired tasks once it owns execution. |
-| `stop()` | Stops claiming work, waits (without a time limit) for the phase in flight while still renewing the owner lease, then releases it. No-op when not started. |
+| `stop()` | Stops claiming work, waits (without a time limit) for every phase in flight while still renewing the owner lease, then releases it. No-op when not started. |
 | `createTask(definition, input, {requestId?, userId?})` | Inserts a `pending` task, its `ObsTrace`, and its root span in one transaction. Wakes the runner. The definition must be in the registry. |
 | `resolveInterrupted(id, {action, reason, result?, userId?})` | Resolve an `interrupted` task. See [resolveInterrupted](#resolveinterrupted). |
 | `abort(id, {reason, userId?})` | Abort a task and every non-terminal task it owns, bottom-up. See [abort](#abort). |
@@ -166,7 +166,7 @@ some), throws. A soft-deleted task still owns its `requestId`.
 | `models` | `({provider, modelId}) => LanguageModel` | Resolves an agent's model refs to Vercel AI SDK models. Required when `registry` lists an agent. |
 | `env` | `ExecutionEnv` | Handed to phases as `rt.env` and to tools as `api.env`. Optional. |
 | `priceMap` | `Record<modelId, {inputPerMTok, outputPerMTok}>` | Prices LLM spans (`usage.costUsd`). Defaults to the registered `ObservabilityApp`'s `priceMap`, read at call time. |
-| `runner` | `HarnessRunner` | Default `new InProcessRunner()`. |
+| `runner` | `HarnessRunner` | Default `new InProcessRunner()`. For many instances, `JobsRunner` (see [JobsRunner](#jobsrunner)). |
 | `streaming` | `{deltaFlushChars?, deltaFlushInterval?, deltaTtl?}` | How streamed model text becomes `delta` events. Defaults: 200 characters, `{milliseconds: 250}`, `{hours: 1}`. Non-positive values throw. See [Event stream (SSE)](#event-stream-sse). |
 | `testHooks` | `HarnessTestHooks` | Test-only. See [Testing](#testing). |
 
@@ -183,7 +183,8 @@ const harness = await Harness.open({
 
 | Option | Default | Description |
 | --- | --- | --- |
-| `pollInterval` | `{milliseconds: 250}` | Idle sleep between claim attempts. `createTask` wakes the runner early. |
+| `concurrency` | `8` | Most claimed tasks run at once. Must be a positive integer, or the constructor throws. `1` runs one task at a time. |
+| `pollInterval` | `{milliseconds: 250}` | Idle sleep between claim attempts. `createTask` and a task that frees a slot wake the runner early. |
 | `leaseDuration` | `{seconds: 30}` | Lifetime of the owner lease and of each task lease without a renewal. Luxon `DurationLike`. |
 | `heartbeatInterval` | `{seconds: 10}` | How often both leases are renewed. Must be positive and shorter than `leaseDuration`, or the constructor throws. |
 | `ownerId` | `hostname:pid:uuid` | Id written into every lease this runner holds. |
@@ -197,13 +198,101 @@ Only the owner claims work. It claims the oldest `pending` task whose `name@vers
 registered, whose `runAt` is empty or past (by Luxon's `DateTime.now()`), and that has no
 `abortRequested`. It sets the task `running` with a fresh task lease and runs phases one
 after another until the task stops: terminal, `pending` for a retry, `waiting`, or
-aborted. One task runs at a time. An idle owner re-polls every `pollInterval`, so a retry
-starts within one poll of its `runAt`.
+aborted. An idle owner re-polls every `pollInterval`, so a retry starts within one poll of
+its `runAt`.
+
+Up to `concurrency` tasks run at once, each under its own task lease and heartbeat, so a
+slow phase (a long model call, a phase blocked on I/O) does not delay other tasks. Ordering
+guarantees:
+
+- Tasks are **claimed** oldest first (`created`), one claim at a time, while a slot is free.
+- With `concurrency` above 1, claimed tasks run in parallel, so a later task can finish
+  first. Use `concurrency: 1` for strict one-at-a-time order.
+- Phases of one task always run in order, and a phase never commits twice: the claim is an
+  atomic update, and every commit is fenced on the task lease. (A runner that froze past
+  its lease may still be executing phase code while recovery replays it; only its commit
+  is rejected.)
+- Turns of one conversation stay serial regardless of `concurrency`; turns of different
+  conversations, and sibling child tasks, run in parallel.
+- `stop()` stops claiming and waits for every task in flight.
+
+Before Phase 2 the runner ran one task at a time. Upgrading to the default of 8 means
+parallel model calls (watch provider rate limits and cost) and more stream-counter retries
+(see [Implementation notes](#implementation-notes)). Pass `concurrency: 1` to keep the old
+behavior.
+
+### JobsRunner
+
+Runs phases as `@terreno/jobs` jobs so any number of instances share the work. Import it
+from its own entry point; `@terreno/jobs` is an optional peer dependency, and importing
+`@terreno/ai` or `@terreno/ai/harness` never loads it. How-to:
+[Run the harness on multiple instances](../how-to/run-harness-on-multiple-instances.md).
+
+```typescript
+import {JobsApp} from "@terreno/jobs";
+import {Harness} from "@terreno/ai/harness";
+import {JobsRunner} from "@terreno/ai/harness/jobsRunner";
+
+const jobs = new JobsApp();
+const harness = await Harness.open({registry, runner: new JobsRunner({jobs})});
+app.register(jobs).build(); // the runner enqueues through the registered jobs service
+await harness.start();
+await jobs.startWorker();
+```
+
+| Option | Default | Description |
+| --- | --- | --- |
+| `jobs` | required | The `JobsApp`. The constructor defines `terreno.harness.phase` on it (`HARNESS_PHASE_JOB_NAME`). Register it and build the app before `harness.start()`; otherwise `start()` throws `harness-config-invalid`. |
+| `pollInterval` | `{milliseconds: 500}` | Sleep between dispatch scans. `createTask`, `sendEvent`, and each finished phase wake the dispatcher early. |
+| `leaseDuration` | `{seconds: 30}` | Lifetime of each task lease without a renewal. |
+| `heartbeatInterval` | `{seconds: 10}` | How often a running phase renews its task lease, and how often this instance sweeps expired leases. Must be positive and shorter than `leaseDuration`, or the constructor throws. |
+| `ownerId` | `hostname:pid:uuid` | Id written into every task lease this runner holds. |
+| `dispatchBatchSize` | `100` | Most runnable tasks one scan enqueues. Must be a positive integer, or the constructor throws. |
+
+Every instance runs a dispatcher and a jobs worker. There is no owner lease.
+
+1. **Dispatch.** Each scan lists runnable tasks (the same rule `InProcessRunner` claims by)
+   and enqueues one `terreno.harness.phase` job per task with payload `{taskId}` and
+   idempotency key `taskId:phase:attempt:claims`. `claims` counts how often the task was
+   claimed, so each runnable visit of a phase (first run, retry, event wake, replay) gets
+   its own job, and repeated scans or instances enqueue it once. When the job with that key
+   already ended without claiming the task (for example it died after its retries, or ran
+   on an instance mid-deploy that did not register the task), the dispatcher follows or
+   extends the chain `<key>:<endedJobId>` until a job is in flight, so the task is not
+   stranded.
+2. **Run one phase.** The job handler claims that task under a fresh fenced task lease,
+   runs **one** phase with heartbeats, and hands a task that is still `running` back as
+   `pending` (lease cleared) for the next job. It then wakes the dispatcher. A handler
+   whose task is no longer runnable (another job claimed it, or it is aborting) does
+   nothing and succeeds.
+3. **Recover.** Each instance sweeps expired task leases once per `heartbeatInterval`, with
+   the same replay and interruption rules as `InProcessRunner`. A replayed task becomes
+   runnable with a new `claims` value, so it is dispatched again.
+
+Load: with N instances, the recovery sweep (expired leases, settled child waits, stranded
+queued conversations) runs N times per `heartbeatInterval`, where `InProcessRunner` runs it
+once, and each scan does one job lookup per runnable task (up to `dispatchBatchSize`). The
+sweeps are fenced, so this costs queries, not correctness. Raise `heartbeatInterval` and
+`pollInterval` for large fleets.
+
+The task lease, not the job lock, is the authority. Two jobs for one phase race on the
+atomic claim, and only one runs it. A worker that froze past its lease cannot commit: its
+`rt.commit` throws `HarnessCommitConflictError`. A phase that runs longer than the jobs
+lock TTL is safe for the same reason. A job that runs on an instance whose harness is not
+started (or is stopping) throws `harness-runner-stopped`, so the jobs worker retries it.
+
+`stop()` stops dispatching and waits for every phase this instance is running. Stop the
+jobs worker (`jobs.stopWorker()`) after `harness.stop()`.
+
+### Custom runners
 
 A custom runner implements `HarnessRunner` (`start(context)`, `stop()`, `wake()`). The
 `context` provides `acquireOwnerLease(lease)`, `releaseOwnerLease(lease)`,
-`recoverExpired()`, `claimNext(lease)`, and `runTask(task, lease)`, where `lease` is
-`{owner, duration, heartbeat}` (Luxon `Duration`s).
+`recoverExpired()`, `claimNext(lease)`, `claimTask(taskId, lease)` (claim one task by id
+when it is runnable now, else `null`), `listRunnable(limit)` (oldest runnable tasks as
+`{taskId, phase, attempt, claims}`, unclaimed), and `runTask(task, lease, {maxPhases?})`,
+where `lease` is `{owner, duration, heartbeat}` (Luxon `Duration`s). With `maxPhases`,
+`runTask` hands a task that is still `running` after that many phases back as `pending`.
 
 ## Runtime (rt)
 
@@ -261,7 +350,9 @@ Assumptions:
   recovery treats them as expired.
 
 Recovery runs when a runner becomes owner (on `start()` or on takeover) and on every owner
-heartbeat. It scans up to 100 `running` tasks whose lease expired (or that have no lease),
+heartbeat. `JobsRunner` has no owner lease: every instance sweeps once per
+`heartbeatInterval`, and concurrent sweeps are fenced, so each expired task is recovered
+once. It scans up to 100 `running` tasks whose lease expired (or that have no lease),
 oldest first, for registered `name@version`s only. A stopping owner skips the sweep. A task
 whose interruption commit keeps failing for a non-conflict reason is logged and retried on
 the next sweep.
@@ -1503,6 +1594,7 @@ empty objects are kept (`minimize: false`), so an initial state `{}` is stored a
 | `outcome` | `{status, result, error}` | Set on terminal commit. |
 | `attempt` | Number | Failed attempts of the current phase. Default 0. |
 | `step` | Number | Phase commits so far; names the current phase visit for idempotent child creation. Default 0. |
+| `claims` | Number | Times a runner claimed the task. Numbers each runnable visit; `JobsRunner` puts it in the job idempotency key. Default 0; a missing value reads as 0. |
 | `retry` | `{maxAttempts, backoffMs, maxBackoffMs}` | Copied from the definition. |
 | `ownership` | `{kind: root \| task \| conversation, id}` | Default `root`. `rt.createTask` children are `{kind: "task", id: <parent>}`; agent turns are `{kind: "conversation", id}`, except subagent turns, which are `{kind: "task", id: <caller>}`. |
 | `rootTaskId` | ObjectId | Top of the ownership tree; equals `_id` for root tasks. |
@@ -1686,12 +1778,13 @@ lists every kind:
 | `code` | `status` | `title` | Kinds of failure |
 | --- | --- | --- | --- |
 | `harness-definition-invalid` | 500 | Invalid harness definition | `defineTask` / `defineTool` / `defineAgent` / `defineExtension` / `approvalGate` validation, and `rt.*` misuse inside a phase. Thrown as `HarnessDefinitionError`, which fails the task at once (no retry). |
-| `harness-config-invalid` | 500 | Invalid harness configuration | `Harness.open`, `HarnessApp`, `InProcessRunner`, `streaming`, and registry problems. |
+| `harness-config-invalid` | 500 | Invalid harness configuration | `Harness.open`, `HarnessApp`, `InProcessRunner` (lease timing, `concurrency`), `JobsRunner`, `streaming`, and registry problems. |
 | `harness-replica-set-required` | 500 | MongoDB replica set required | `Harness.open` on a deployment without transactions. |
-| `harness-invalid-request` | 400 | Invalid harness request | A caller passed a bad argument (`abort`, `sendEvent`, `decideApproval`, `resolveInterrupted`, `submit` / `send`, `createConversation`, memos). |
+| `harness-invalid-request` | 400 | Invalid harness request | A caller passed a bad argument (`abort`, `sendEvent`, `decideApproval`, `resolveInterrupted`, `submit` / `send`, `createConversation`, memos, a `terreno.harness.phase` payload without `taskId`). |
 | `harness-not-registered` | 404 | Not registered in this harness | A task definition, agent, or extension the registry lacks. |
 | `harness-not-found` | 404 | Harness record not found | A conversation or child task disappeared. |
-| `harness-already-started` | 409 | Already started | `Harness.start` / `InProcessRunner.start` called twice. |
+| `harness-already-started` | 409 | Already started | `Harness.start` / `InProcessRunner.start` / `JobsRunner.start` called twice. |
+| `harness-runner-stopped` | 503 | Harness runner is not running | A `terreno.harness.phase` job ran on an instance whose harness is not started; the jobs worker retries it. |
 | `harness-commit-conflict` | 409 | Harness commit lost its checkpoint fence | `HarnessCommitConflictError`. |
 | `harness-approval-not-pending` | 409 | Approval can no longer be decided | `HarnessApprovalConflictError`. |
 | `harness-conversation-busy` | 409 | Conversation is busy | `HarnessConversationBusyError`. |
@@ -1809,7 +1902,7 @@ Low-risk choices made in the first slice:
 | The SSE tail starts at a cluster time read before any subscriber's replay query | No gap between replay and tail; overlap is removed by `seq`. |
 | One shared change stream per `HarnessApp`, not one per connection | A change stream holds a pooled connection while it waits; per-connection streams let a few dozen viewers starve commits and lease renewals. |
 | Queue recovery is a sweep by the execution owner, not part of the turn's terminal commit | Starting a turn is its own transaction; the sweep (fenced, idempotent) covers the crash window between the two. |
-| Writers to one stream conflict on its counter document | The price of commit-ordered `seq`s. `withTransaction` retries the loser. `InProcessRunner` runs one task at a time, so only HTTP submits, approvals, and deltas contend; a parallel runner fanning out many children of one tree will see retries. |
+| Writers to one stream conflict on its counter document | The price of commit-ordered `seq`s. `withTransaction` retries the loser. `InProcessRunner` (default `concurrency: 8`) and `JobsRunner` both run sibling tasks of one tree in parallel, so their writes to the shared root stream retry alongside HTTP submits, approvals, and deltas. `InProcessRunner({concurrency: 1})` avoids task-on-task contention. |
 | Interruption spans start at `lease.acquiredAt` | Records when the cut-off phase began. |
 | `resolveInterrupted({action: "abort"})` closes the trace as `error` | `ObsTrace.status` is `ok` or `error`; an abort is not a success. |
 | `requestId` replays are scoped to the same `userId` | Stops one caller from reading another user's task through a shared key. |

package/src/docs/versioned/next/reference/api.md

@@ -41,7 +41,7 @@ REST API framework built on Express and Mongoose. Provides modelRouter (CRUD end
 - Validation: `configureOpenApiValidator`, `validateRequestBody`, `validateQueryParams`, `createValidator`
 - Middleware: `openApiEtagMiddleware`, `sentryAppVersionMiddleware`
 - Extensibility: `TerrenoPlugin` interface
-- Notifiers: `sendToSlack`, `formatSlackUserMention`, `lookupSlackUserIdByEmail`, `sendToGoogleChat`, `sendToZoom`
+- Notifiers: `sendToSlack`, `formatSlackUserMention`, `lookupSlackUserIdByEmail`, `createSlackPrivateChannel`, `inviteSlackUsersToChannel`, `postSlackMessage`, `findSlackChannelByName`, `sendToGoogleChat`, `sendToZoom`
 - HTTP client: `createAuthenticatedClient`, `withApiErrorHandling`, `normalizeApiError`, `markRetryUnsafe`
 
 ## Server Setup
@@ -224,6 +224,30 @@ export const todoRouter = modelRouter("/todos", Todo, {
 Do not add `endpoints: (router) => { router.get(...) }` when an action fits.
 Action config supports `allowAnonymous: true` as a per-action override; it does not change authentication for CRUD or sibling actions.
 
+## modelRouter list queries
+
+`GET /{path}` accepts filters on fields listed in `queryFields` (qs syntax, so operators work:
+`?status[$in][0]=open&created[$gte]=2026-01-01`), plus the reserved `limit`, `page`, and `sort`.
+`limit` is clamped to `maxLimit` (default 500); the response echoes the applied `limit` with
+`{data, limit, more, page, total}`.
+
+A filter on a field missing from `queryFields` is rejected before query validation runs, so
+it can never be stripped and silently ignored:
+
+```json
+{
+  "status": 400,
+  "code": "query-param-not-allowed",
+  "title": "Query parameter not allowed",
+  "detail": "title is not allowed as a query param. Add \"title\" to queryFields on the Todo modelRouter to filter on it. Allowed: completed, created, ownerId.",
+  "source": {"parameter": "title"},
+  "meta": {"queryParam": "title", "model": "Todo", "allowedQueryFields": ["completed", "created", "ownerId"]}
+}
+```
+
+`@terreno/syncdb` raises this as `QueryFieldNotAllowedError` (see
+[Query windows](syncdb.md#query-windows)).
+
 ## modelRouter array operations
 
 When a model has at least one array field, `modelRouter` provides subroutes for adding,
@@ -1581,6 +1605,44 @@ await sendToSlack(`${formatSlackUserMention("U012ABCDEF")} Deployment complete`,
 });
 ``````
 
+Incoming webhooks are fixed to the channel chosen when the webhook was created.
+Private channels, invites, and messages to a channel id use the bot token for
+one Slack workspace (`SLACK_BOT_TOKEN`):
+
+| Function | Slack method | Bot scopes |
+| --- | --- | --- |
+| `createSlackPrivateChannel` | `conversations.create`, `conversations.invite`, `chat.postMessage` | `groups:write`, `groups:write.invites`, `chat:write` |
+| `inviteSlackUsersToChannel` | `conversations.invite` | `groups:write.invites` for private channels, `channels:write.invites` for public |
+| `postSlackMessage` | `chat.postMessage` | `chat:write` |
+| `findSlackChannelByName` | `conversations.list` | `channels:read`, `groups:read` |
+| `lookupSlackUserIdByEmail` | `users.lookupByEmail` | `users:read.email` |
+
+The bot can post and invite only in channels it belongs to. Creating a private
+channel makes the bot a member. An existing private channel must already
+include the app; a bot cannot join a private channel by itself. Store the
+returned channel id. Names are unique, mutable, and limited to lowercase
+letters, numbers, hyphens, and underscores.
+
+``````typescript
+import {createSlackPrivateChannel, inviteSlackUsersToChannel} from "@terreno/api";
+
+const channel = await createSlackPrivateChannel({
+  name: "case-12345",
+  userIds: ["U012ABCDEF"],
+  initialMessage: "Case opened.",
+  mentionUserIds: ["U012ABCDEF"],
+});
+
+await inviteSlackUsersToChannel({
+  channelId: channel.channelId,
+  userIds: ["U098765432"],
+});
+``````
+
+A taken name throws status 409. `findSlackChannelByName({name})` returns a
+public channel, or a private channel the bot is already in. It returns
+`undefined` when that channel is not visible to the bot.
+
 ### Google Chat Notifications
 
 ``````typescript
@@ -1913,7 +1975,7 @@ Complete reference of environment variables used by @terreno/api:
 | Variable | Required | Default | Description |
 |----------|----------|---------|-------------|
 | `SLACK_WEBHOOKS` | No | — | JSON object mapping names to Slack webhook URLs: `{"default":"https://..."}` |
-| `SLACK_BOT_TOKEN` | No | — | Bot token (`xoxb-…`) with `users:read.email` for `lookupSlackUserIdByEmail` |
+| `SLACK_BOT_TOKEN` | No | — | Bot token (`xoxb-…`) for one workspace. Scopes: `users:read.email`, `groups:write`, `groups:write.invites`, `channels:write.invites`, `chat:write`, `channels:read`, `groups:read` |
 | `GOOGLE_CHAT_WEBHOOKS` | No | — | JSON object mapping names to Google Chat webhook URLs |
 | `ZOOM_CHAT_WEBHOOKS` | No | — | JSON object mapping names to Zoom webhook URLs |
 | `WEBHOOK_SECRET` | No | — | Secret for validating incoming webhook signatures |

package/src/docs/versioned/next/reference/environment-variables.md

@@ -223,7 +223,7 @@ There is **no** `RATE_LIMIT_ENABLED` (or similar) read by `@terreno/api`. Apps t
 |----------|---------|----------|---------|--------|-------|
 | `SLACK_WEBHOOKS` | `@terreno/api` | ❌ | — | Yes | server |
 | `SLACK_WEBHOOK_URL` | scripts | ❌ | — | Yes | server |
-| `SLACK_BOT_TOKEN` | `@terreno/api` `lookupSlackUserIdByEmail` | ❌ | — | Yes | server |
+| `SLACK_BOT_TOKEN` | `@terreno/api` Slack Web API (`lookupSlackUserIdByEmail`, private channels, invites, `chat.postMessage`) | ❌ | — | Yes | server |
 | `GOOGLE_CHAT_WEBHOOKS` | `@terreno/api` | ❌ | — | Yes | server |
 | `GOOGLE_CHAT_WEBHOOK_URL` | scripts | ❌ | — | Yes | server |
 | `ZOOM_CHAT_WEBHOOKS` | `@terreno/api` | ❌ | — | Yes | server |

package/src/docs/versioned/next/reference/lifecycle-plugin.md

@@ -16,6 +16,17 @@ loops that invoke those stages; they are not stages and must not appear as `stag
 | `terreno-4-brew` | Roast PASS for every in-scope task + branch/evidence | pushed head + PR + product-CI trigger check + review-bot wait + attached evidence | Taste |
 | `terreno-5-taste` | PR + current state | one current-head reaction; before push: fetch latest `master` (merge only when needed), re-verify last-run failed tests locally, run root `prepush` when present (otherwise affected-package checks) in a no-context subagent, then watch CI; invoked directly, repeats bounded reactions until terminal | null or fresh Taste |
 
+Pre-Grow skill (not a stage):
+
+| Skill | Input | Primary output | Next |
+| --- | --- | --- | --- |
+| `terreno-prd` | app owner + conversation, prototype, repository | interview, then a published full product PRD (problem, why now, solution, user stories, success metrics and guardrails, constraints, rollout, out of scope); every metric measurable or `TBD` with its decider | Grow, with the PRD as its spec; metrics become acceptance criteria, `TBD`s become grilling questions, settled product decisions are not re-asked |
+
+It is explicit-invoke only (`/terreno-prd`, `$terreno-prd`, `/terreno:prd`). It
+interviews with Grow's grilling procedure over a product tree, is sized for one app owner
+and one or two developers, and carries no implementation or testing decisions; those stay
+in Grow.
+
 Outer loops (not stages):
 
 | Skill | What it walks | Default |

package/src/docs/versioned/next/reference/syncdb.md

@@ -10,6 +10,7 @@ Local-first data layer for Terreno frontends. A TinyBase `MergeableStore` on dev
 - [createSyncDb configuration](#createsyncdb-configuration)
 - [SyncDb client methods](#syncdb-client-methods)
 - [React hooks](#react-hooks)
+- [Query windows](#query-windows)
 - [Codegen](#codegen)
 - [Conflict API](#conflict-api)
 - [Sync status API](#sync-status-api)
@@ -25,7 +26,8 @@ Local-first data layer for Terreno frontends. A TinyBase `MergeableStore` on dev
 - `createSyncDb`, `SyncDb`, `SyncDbConfig`, `MutateArgs`
 - `betterAuthAdapter`, `bridgeBetterAuthReactClient`, `AuthProvider`
 - `listConflicts`, `wipeLocalData`, `generateMutationId`
-- React (`@terreno/syncdb/react`): `SyncDbProvider`, `useEntity`, `useQuery`, `useEntityIds`, `useMutate`, `useSyncStatus`, `useConflicts`, `useSyncDebugLog`, `createCollectionHooks`
+- Queries: `QueryFieldNotAllowedError`, `ListRequestError`, `WhereFilter`, `SortSpec`, `WindowQuery`, `QueryWindowState`, `QueryWindows`, `compileWhere`, `DEFAULT_WINDOW_PAGE_SIZE`
+- React (`@terreno/syncdb/react`): `SyncDbProvider`, `useEntity`, `useQuery`, `useEntityIds`, `useWindowQuery`, `useMutate`, `useSyncStatus`, `useConflicts`, `useSyncDebugLog`, `createCollectionHooks`
 - CLI: `terreno-syncdb-codegen` (generates `SYNC_COLLECTIONS` + friendly hooks from OpenAPI)
 - Testing (`@terreno/syncdb/testing`): `createFakeTransport`
 
@@ -165,6 +167,7 @@ be passed to `betterAuthAdapter` directly.
 | `name` | `string` | — (required) | Persisted database name |
 | `collections` | `string[]` | — (required) | Collection names to sync (local tables + subscriptions) |
 | `windowCollections` | `string[]` | `[]` | Collections that join `{collection}\|admin`, skip snapshots/reconcile, and hydrate only known REST membership ids |
+| `queryCollections` | `Array<string \| {collection, path?, fullSync?}>` | `[]` | Collections too large to sync whole (or `fullSync: true` to keep snapshot sync and add windows). No snapshot/reconcile paging; rows arrive through [query windows](#query-windows) against the `modelRouter` list endpoint (`path`, default `/{collection}`) plus live deltas on the user's normal streams. Each entry must also be in `collections` and not in `windowCollections`. |
 | `organizationIdProvider` | `() => string \| undefined` | — | Read at send time: sets `X-Organization-Id` on HTTP sync calls and `organizationId` on socket mutate payloads |
 | `authProvider` | `AuthProvider` | — (required) | `{getToken, getUserId, onAuthChange, refresh?}` |
 | `baseUrl` | `string` | — | Server origin; required unless both `transport` and `httpChannel` are injected |
@@ -200,6 +203,7 @@ be passed to `betterAuthAdapter` directly.
 | `mutate({collection, operation, id?, data?})` | Optimistic local write + durable outbox enqueue + fire-and-forget replay. Returns `{mutationId, id}`. |
 | `reconcile()` | HTTP snapshot catch-up for every known stream; runs tombstone compaction on success. Also runs automatically on (re)connect, on a rate-limited seq-jump hint, and on the periodic timer; each `sync:subscribed` confirmation additionally pages just the streams it names. |
 | `hydrateWindow({collection, ids, restRows?})` | Admin window upsert: REST rows (optional) land immediately; every requested id is also fetched from `GET /sync/entities` before this resolves so seq/deleted metadata is canonical for immediate update/delete. A `{collection}|admin` delta that lands while the fetch is in flight wins, so hydration never rewinds a row to an older seq. Unknown ids are ignored. |
+| `queryWindows` | Server query windows for `queryCollections`: `fetchWindow({query, nextPage?})`, `getWindow(query)`, `retain(query)` → release, `refetchRetained()`, `pruneUnretained()`, `subscribe(cb)`, `keyFor(query)`. See [Query windows](#query-windows). |
 | `forceResync()` | Purge every known stream locally and re-bootstrap from cursor 0 (outbox/conflicts untouched). Returns `{ok, reason?, streams, purged, repaired}`. After discovery and each stream bootstrap it waits for in-flight `start`/`stop`/auth-change work, then abandons with `reason: "superseded"` when that work switched users or generations. |
 | `replayOutbox()` | Drain queued mutations for the current user now. |
 | `resolveConflict({mutationId, strategy})` | Apply `"useServer"` or `"keepMine"` to a recorded conflict. |
@@ -271,13 +275,26 @@ Subscribe to a single entity. Re-renders when that row changes. `isPending` is t
 
 ```typescript
 const todos = useQuery<Todo>("todos", {
-  filter: (t) => !t.completed,
-  sort: (a, b) => b.created.localeCompare(a.created),
-  includeDeleted: false,
+  where: {completed: false, created: {$gte: since}},
+  sort: "-created",
+  limit: 20,
 });
+
+// Callbacks still work and combine with `where`:
+const mine = useQuery<Todo>("todos", {filter: (t) => t.ownerId === userId, sort: byCreatedDesc});
 ```
 
-Returns decoded entity data arrays. Tombstones excluded unless `includeDeleted: true`. Filter and sort run in JS — memoize callbacks when collections are large.
+Returns decoded entity data arrays from the local store. Tombstones excluded unless `includeDeleted: true`.
+
+| Option | Type | Description |
+|--------|------|-------------|
+| `where` | `WhereFilter` | Mongo-style filter: equality, dot paths, `$eq` `$ne` `$gt` `$gte` `$lt` `$lte` `$in` `$nin` `$exists`, `$and` / `$or`. An array field matches when any element equals the value. Compared by value, so inline objects are fine. Throws for operators only the server can run (`$search`, `$regex`). |
+| `filter` | `(data) => boolean` | Arbitrary predicate, applied after `where`. |
+| `sort` | `SortSpec \| comparator` | `"-created name"`, `{created: "descending"}`, or `(a, b) => number`. |
+| `limit` | `number` | Keep the first N rows after filter and sort. |
+| `includeDeleted` | `boolean` | Include tombstones. |
+
+Filter and sort run in JS — memoize `filter`/comparator callbacks when collections are large.
 
 ### `useEntityIds(collection, options?)`
 
@@ -288,6 +305,38 @@ return ids.map((id) => <TodoRow key={id} id={id} />);
 
 Same options as `useQuery`, but returns only ordered ids with **referential stability** — the array identity changes only when membership or order changes, not on field updates. Pair with per-row `useEntity` for large lists.
 
+### `useWindowQuery(collection, options?)`
+
+```typescript
+const {data, isLoading, isFetching, hasMore, fetchNextPage, refetch, total, error} =
+  useWindowQuery<Message>("messages", {
+    where: {threadId, deleted: false},
+    sort: "-created",
+    pageSize: 30,
+  });
+```
+
+Server-filtered list over a `queryCollections` collection. See [Query windows](#query-windows).
+
+| Option | Type | Default | Description |
+|--------|------|---------|-------------|
+| `where` | `WhereFilter` | — | Sent as qs list params (`status[$in][0]=open`). Fields must be in the route's `queryFields`. |
+| `sort` | `SortSpec` | route default | Server `sort`; also orders the local view. |
+| `pageSize` | `number` | `50` | Server `limit` per page (capped by the route's `maxLimit`). |
+| `skip` | `boolean` | `false` | Do not retain or fetch (e.g. a required param is missing). |
+| `liveMatch` | `boolean` | `true` | Show cached rows that newly match `where` inside the loaded range. |
+
+| Result | Description |
+|--------|-------------|
+| `data` / `ids` | This window's rows in view order. |
+| `isLoading` | No membership yet (first fetch in flight). Cached membership renders with `isLoading: false`. |
+| `isFetching` | Any request for this window is in flight. |
+| `isError` / `error` | Last request failed (cached rows still render). |
+| `errorCode` | Server error code of the last failure; `"query-param-not-allowed"` means a `where` field is missing from `queryFields`. |
+| `hasMore` / `total` | From the list response's `more` / `total`. |
+| `fetchNextPage()` | Append the next server page. |
+| `refetch()` | Reload every loaded page in one request, replacing membership. |
+
 ### `useMutate(collection)`
 
 ```typescript
@@ -331,7 +380,7 @@ Returns the `SyncDb` instance from context (escape hatch for imperative calls li
 
 ### `createCollectionHooks`
 
-Factory used by `terreno-syncdb-codegen` and by hand-written custom collections. Returns five operation hooks (`useListQuery`, `useReadQuery`, `useCreateMutation`, `useUpdateMutation`, `useDeleteMutation`). Generated SDKs rename them to friendly names (`useTodos`, `useTodo`, `useCreateTodo`, …). Mutation hooks return `[trigger]`; triggers apply locally and return `{mutationId, id}` synchronously. Optional `retries` maps to `maxAttempts` (`false` → 1, a number → that many, omitted → engine default).
+Factory used by `terreno-syncdb-codegen` and by hand-written custom collections. Returns six operation hooks (`useListQuery`, `useWindowQuery`, `useReadQuery`, `useCreateMutation`, `useUpdateMutation`, `useDeleteMutation`). Generated SDKs rename them to friendly names (`useTodos`, `useTodosWindow`, `useTodo`, `useCreateTodo`, …). `useWindowQuery` works only for collections listed in `queryCollections`. Mutation hooks return `[trigger]`; triggers apply locally and return `{mutationId, id}` synchronously. Optional `retries` maps to `maxAttempts` (`false` → 1, a number → that many, omitted → engine default).
 
 ```typescript
 export const {useListQuery: useNotes, useCreateMutation: useCreateNote} =
@@ -340,6 +389,107 @@ export const {useListQuery: useNotes, useCreateMutation: useCreateNote} =
   });
 ```
 
+## Query windows
+
+Use query windows when a collection is too large to sync whole (messages, events, audit
+rows). They match what an RTK list query did — `where`, `sort`, `limit`, `page`, `more`,
+`total` — but keep the data local-first and deduplicated.
+
+```typescript
+export const syncDb = createSyncDb({
+  collections: ["todos", "messages"],
+  queryCollections: [{collection: "messages", path: "/messages"}],
+  // ...
+});
+```
+
+**Membership, not copies.** A window is one list query against the collection's
+`modelRouter` list endpoint. Its result is an ordered id list stored in the reserved
+`_queryWindows` table. Each row lands once in the collection's entity table, keyed by id.
+Two windows that overlap share those rows:
+
+- A delta, mutation, or refetch for a shared row writes the store once; every window that
+  lists the row re-renders from it.
+- Each window lists only its own members. Window B never shows rows that only window A
+  fetched.
+
+**Live membership.** When `where` is locally evaluable (see `useQuery`'s `where`), a window
+also:
+
+- hides a member whose fields stop matching (an unpinned message leaves the "pinned" window);
+- shows cached rows that newly match (a live delta, or a local create) when they sort inside
+  the loaded range. With more pages unloaded, a row that sorts after the last loaded row
+  waits for `fetchNextPage()`. Without a `sort` and with more pages, only local creates are
+  admitted.
+
+A full fetch also records cached rows that matched `where` locally but that the server did
+not return (deleted, changed, or no longer visible while this device was offline). They stay
+out of the view until a newer delta or a local edit changes them, so stale cached data never
+resurrects a row.
+
+Set `liveMatch: false` to show server membership only. Operators only the server runs
+(`$search`) also fall back to server membership.
+
+**Fetching.** The first `retain` of a window in a session fetches it; persisted membership
+renders meanwhile, so windows work offline. Concurrent fetches of the same window share one
+request; a refetch and a `fetchNextPage()` on one window run one after the other. Retained
+windows refetch on reconnect and after `forceResync()` — query collections
+skip snapshot catch-up, so this is how deltas missed offline are recovered. Windows not on
+screen at that moment refetch the next time they are shown. Window fetches
+started before a `stop()` or user switch are discarded.
+
+**Seq.** List rows that carry `_syncSeq` (the `syncPlugin` field) are written with that seq.
+Rows without it (a `responseHandler` that strips it) are re-read from `GET /sync/entities` so
+a later mutation never sends a stale `baseVersion`. Rows with a pending outbox mutation, or
+with a newer seq from a delta, are never overwritten. A list `responseHandler` must return the
+same shape as the sync serializer; otherwise list rows and delta rows differ.
+
+**Missing `queryFields`.** Every `where` field must be in the route's `queryFields`. The
+server answers 400 `query-param-not-allowed` naming the field, the model, and the allowed
+list. syncdb turns it into `QueryFieldNotAllowedError` (`field`, `allowedQueryFields`,
+`model`), logs it with `console.error`, sets the window's `errorCode` to
+`"query-param-not-allowed"`, and rejects `fetchWindow` / `refetch` / `fetchNextPage` with it.
+Other list failures throw `ListRequestError` (`status`, `code`, `detail`).
+
+```typescript
+import {QueryFieldNotAllowedError} from "@terreno/syncdb";
+
+try {
+  await refetch();
+} catch (error) {
+  if (error instanceof QueryFieldNotAllowedError) {
+    // error.field === "title", error.allowedQueryFields === ["completed", "created", "ownerId"]
+  }
+}
+```
+
+**`maxLimit` and pipelining.** `modelRouter` clamps `limit` to the route's `maxLimit` and
+echoes the applied value. Window requests always send `page` (so the server never logs a
+truncated unpaginated list). When a fetch needs more rows than one request returns — a
+`refetch()` after several `fetchNextPage()` calls, or a `pageSize` above `maxLimit` — syncdb
+learns the cap from the echoed `limit` and splits the range into sequential page requests
+of `maxLimit` rows starting at the aligned page that contains the range, trimming the extra
+rows, so offsets stay exact. Each pipelined fetch logs a
+`console.warn` naming the row count, the cap, and the request count. Fix it by loading fewer
+pages, lowering `pageSize`, or raising `maxLimit` on the route.
+
+**Fully synced collections.** `queryCollections: [{collection: "todos", fullSync: true}]`
+keeps normal snapshot sync and also enables windows, for server-side filtering, paging, and
+totals over a collection that is already local. See `example-frontend/app/todo-windows.tsx`.
+
+**Limits.**
+
+- `refetch()` with several pages loaded reloads `pageSize × pages` rows (see below).
+- Page appends dedupe ids. A delete between page loads can shift offsets and skip one row
+  until the next refetch.
+- `null` in `where` is rejected (qs would send the string `"null"`); use `{$exists: false}`.
+  A top-level `{$in: []}` returns an empty window without a request; an empty array anywhere
+  else throws instead of silently widening the query.
+- Rows stay in the entity table after their windows unmount. `start()` prunes persisted
+  membership of unretained windows not fetched for 7 days (`pruneStale`);
+  `pruneUnretained()` drops all unretained membership. Rows are dropped by leave-purge,
+  tombstone compaction, or a wipe.
+
 ## Codegen
 
 `terreno-syncdb-codegen` is a bin of `@terreno/syncdb`. It reads OpenAPI, discovers list operations with `x-terreno-sync`, and writes typed hooks plus `SYNC_COLLECTIONS`.

package/src/docs/versioned/next/reference/ui.md

@@ -725,11 +725,12 @@ Master-detail layout. Pass `listViewData` plus `renderListViewItem` for the list
 screens the detail replaces the list until the user goes back.
 
 `desktopChildrenMinWidth` opts into a minimum pixel width for each desktop child when there
-are one or two children. `narrowViewportChildLabels` opts into a labeled full-width pager on
-the narrow viewport. `narrowBelowWidth` uses that viewport when the window is at or below the
-given pixel width; when omitted, the narrow viewport follows `isNarrowViewport()`. These props
-are web only; the native `SplitPage` ignores them. See `SplitPageProps` for when each prop
-applies and what is ignored.
+are one or two children. `childColumnRounding` sets the border radius on each detail child
+column container (defaults to `md`, 4px); the list column is excluded. `narrowViewportChildLabels`
+opts into a labeled full-width pager on the narrow viewport. `narrowBelowWidth` uses that
+viewport when the window is at or below the given pixel width; when omitted, the narrow
+viewport follows `isNarrowViewport()`. These props are web only; the native `SplitPage`
+ignores them. See `SplitPageProps` for when each prop applies and what is ignored.
 `IconButton`'s `backgroundOpacity` tints only that button's background.
 
 ```typescript

package/src/docs/versioned/next/tasks/durable-agent-harness.md

@@ -131,21 +131,23 @@ created in 1.1 and extended by every later task — never deferred.
 
 ## Phase 2 — Multi-instance
 
-- [ ] **Task 2.0**: `InProcessRunner` concurrency
+- [x] **Task 2.0**: `InProcessRunner` concurrency
   - Delivers: `new InProcessRunner({concurrency})` (default 8) runs up to N claimed tasks at once, each with its own lease and heartbeat; a slow phase no longer delays other tasks' wakes; `stop()` waits for all in-flight phases
   - Files: `ai/src/harness/runners/inProcessRunner.ts`, tests
   - Blocked by: 1.12
   - Skills: `terreno-backend-api`
   - Docs: reference "Runners" (concurrency option, ordering guarantees)
   - Acceptance: bun test — a phase blocked on a gate does not prevent a second task from completing; never more than N concurrent; no task runs twice concurrently
+  - Shipped: fake-context cases in `ai/src/harness/runners/inProcessRunner.test.ts` (default 8, gate, peak, stop waits for all, invalid concurrency) and real-Mongo cases in `ai/src/harness/concurrency.test.ts` (blocked task does not hold up a second; 7 two-phase tasks at concurrency 3 peak at exactly 3 with no task running twice). Roast: PASS (2026-10-04; 76 targeted tests, three mutation checks red)
 
-- [ ] **Task 2.1**: `JobsRunner`
+- [x] **Task 2.1**: `JobsRunner`
   - Delivers: `@terreno/ai/harness/jobsRunner` with optional peer `@terreno/jobs`; job `terreno.harness.phase` keyed `taskId:phase:attempt`; harness task lease + heartbeat is the authority; importing the root `@terreno/ai` never loads jobs
   - Files: `ai/src/harness/runners/jobsRunner.ts`, `ai/package.json`, tests
   - Blocked by: 2.0
   - Skills: `terreno-backend-api`
   - Docs: reference "Runners"; how-to "Run on multiple instances"
   - Acceptance: AC11 — two workers race one phase; exactly one commit; stale owner rejected; knip clean without jobs installed
+  - Shipped: `ai/src/harness/runners/jobsRunner.test.ts` (two-worker race, stale-lease commit rejected, re-dispatch of a dead job, root imports never reach `@terreno/jobs`). Drift: the job key is `taskId:phase:attempt:claims` — `claims` (new `HarnessTask` counter) gives a phase revisited by an event wake or crash replay its own job, since jobs idempotency keys are permanent. Built before Task 2.0; it does not depend on `InProcessRunner` concurrency. Roast: PASS (2026-10-04; 17 targeted tests, ai suite 1185/0, knip clean, two mutation checks red)
 
 - [ ] **Task 2.2**: Runs admin screen
   - Delivers: task list/tree with status filters, abort, resolve-interrupted (retry/abort/complete + reason), link to trace