stitchkit 0.112.3 → 0.113.1

package/llms/cli.2.txt

@@ -149,9 +149,10 @@ One unconvertible schema does not take the connection down. The tool
 is skipped and **named** (`onSkippedTool`, or a stderr line by default), so a
 surface of two hundred tools is not lost to one.
 
-**A discovered schema is checked locally, except for its unions.** The arguments are
-coerced and parsed against the schema the server advertised, so a missing field or a
-wrong scalar is refused before anything is sent. A union in that schema is not judged
+**A discovered schema leaves unions and string formats to its server.** The arguments are
+coerced and parsed against the schema the server advertised. Reported structural failures
+(a missing field, wrong scalar type, unknown key or enum value) and bounds stay local.
+A union in that schema is not judged
 here: a server's presentation may join a discriminated union into one object whose
 shared field is an unlabelled `anyOf` (a stitchkit server with `flattenUnionInput`
 does), and judged against that, a wrong value could only be refused with every branch
@@ -159,6 +160,16 @@ listed. The value goes to the server, which holds the original schema, and its r
 names the one wrong field — `blocks.0.data.artifactId: expected string` rather than
 `No union branch matched — branch 1 … and 6 more branches`.
 
+String formats and patterns also lose the author's error messages in JSON Schema. A
+restored `date-time`, email, UUID or regex check would substitute Zod's generic wording.
+Those `invalid_format` refusals go to the remote instead: a bare `2026-10-12` reaches
+the original datetime schema, whose answer can explain that a full UTC timestamp such
+as `2026-10-12T00:00:00Z` is required. This also applies inside arrays, objects and
+record keys. A mixed refusal retains only its reported local reasons; a record-key
+wrapper stays local when it contains any of those reasons. A schema authored locally
+keeps its complete verdict, including custom errors and refinements. No second parse
+invents checks that the first parse did not reach. See [ADR 0272](../decisions/0272-a-restored-string-format-is-judged-by-its-server.md).
+
 MCP connection limits follow the operation phase. `initialize`, the initialized
 notification, `tools/list` and legacy endpoint discovery use `limits.discovery`;
 `tools/call` uses `limits.call`. A mounted call's handshake still uses discovery

package/llms/stitchkit.txt

@@ -85,12 +85,12 @@ own, recorded as an ADR.
 | `stitchkit/geo` | server (Bun or Node) | evolving | managed GeoIP reader generations, last-known-good reload and the optional MaxMind adapter |
 | `stitchkit/oauth` | browser **and** server | evolving | provider-neutral Authorization Code + PKCE request and one-shot callback transaction mechanics |
 | `stitchkit/google` | server (Bun or Node) | evolving | optional Google code exchange and verified OIDC identity adapter |
-| `stitchkit/observability` | server | stable<br>_redefined in 2 of the 57 minors since 0.56.2, most recently 0.92.0_ | request/tool event projections — `createObservability`, trace context, sanitisation |
+| `stitchkit/observability` | server | stable<br>_redefined in 2 of the 58 minors since 0.56.2, most recently 0.92.0_ | request/tool event projections — `createObservability`, trace context, sanitisation |
 | `stitchkit/testing` | tests on Bun or Node | stable | in-process generated clients over a real Fetch handler, the managed-resource conformance kit and manual time for schedules and retries |
 | `stitchkit/declaration` | browser + build and deployment tooling (Bun or Node) | evolving | `ProjectDeclarationSchema` — the one machine-readable statement a repository makes about itself |
 | `stitchkit/react` | browser + server rendering | stable | `createCursorQuery`, `createCacheBridge`, QueryClient and `ApiError` retry policy |
 | `stitchkit/react/keyboard` | browser | evolving | which part of a screen gets a key — ordered layers, one Escape per level, list zones with real focus; see [keyboard guide](keyboard.md) |
-| `stitchkit/agent-runtime` | server | evolving<br>_redefined in 23 of the 57 minors since 0.56.2, most recently 0.105.0_ | optional durable conversation/run loop, history, models, prompts, fencing and events |
+| `stitchkit/agent-runtime` | server | evolving<br>_redefined in 24 of the 58 minors since 0.56.2, most recently 0.113.0_ | optional durable conversation/run loop, history, models, prompts, fencing and events |
 | `stitchkit/agent-runtime/sandbox` | server | evolving | optional Linux Bubblewrap sessions, durable workspaces and a host HTTP credential gateway; see [sandbox guide](sandbox.md) |
 | `stitchkit/agent-runtime/testing` | tests on Bun or Node | evolving | credential-free replay, scripted provider faults, deterministic race controls and the agent-store conformance kit |
 | `stitchkit/agent-runtime/harness` | server | evolving | resource-aware process-local facade over the canonical Agent runtime; supervision stays outside |
@@ -102,7 +102,7 @@ own, recorded as an ADR.
 | `stitchkit/agent-runtime/sqlite/bun` | server (Bun) | evolving | durable built-in SQLite store for the agent runtime |
 | `stitchkit/agent-runtime/sqlite/node` | server (Node ≥ 22.5) | evolving | durable built-in SQLite store for the agent runtime |
 | `stitchkit-tui` | terminal (Bun) | evolving | optional official OpenTUI host over a caller-composed headless runtime |
-| `stitchkit/application` | browser + server | evolving<br>_redefined in 14 of the 57 minors since 0.56.2, most recently 0.110.0_ | managed resource graph, readiness, admission, schedules, subtree restart and bounded shutdown |
+| `stitchkit/application` | browser + server | evolving<br>_redefined in 14 of the 58 minors since 0.56.2, most recently 0.110.0_ | managed resource graph, readiness, admission, schedules, subtree restart and bounded shutdown |
 | `stitchkit/application/grammy` | server | evolving | a grammY bot as application resources — configuration, then polling with updates admitted by batch — and the webhook adapter |
 | `stitchkit/application/opentelemetry` | server | evolving | maps application snapshots onto an injected OpenTelemetry `Meter` |
 | `stitchkit/application/schemas` | browser + server | evolving | the application's snapshot, health and shutdown schemas alone, without the kernel |

package/llms/tools-connections.txt

@@ -27,7 +27,7 @@ and approval path.
 | `ConnectionPhase` / `ConnectionOperation` / `ConnectionFailureContext` | _type_ | closed operation/phase vocabulary and measured read context for existing typed connection errors |
 | `ConnectionTimeoutError` / `ConnectionResponseTooLargeError` | class | bounded operation refusal, with phase, operation and finite observed bytes; safe tool projection retains the raw cause locally |
 | `defineOpenApiConnection` | function | declare one OpenAPI document (object, JSON text or JSON URL) as a tool surface |
-| `mountConnections` | function | discover every connection and return readonly `RuntimeToolDefinition`s for `mountAgent`; a call's arguments are parsed against the discovered schema before sending, except its unions, which the server judges (ADR 0266) |
+| `mountConnections` | function | discover every connection and return readonly `RuntimeToolDefinition`s for `mountAgent`; reported structural failures and bounds stay local, while unions and string-format/pattern refusals (including inside record keys) are judged by the server that owns the original schema (ADR 0266, ADR 0272) |
 | `ConnectionAuthorizationRequiredError` | class | typed reauthorization signal raised on `401` with credentials resolved again on the next call |
 | `ConnectionBudgetExceededError` | class | the mount would expose more foreign tools than its declared budget |
 | `ConnectionRequestError` | class | any non-`2xx` connection response other than the reauthorization signal |

package/llms/{upgrading-0.103.0-0.112.0.txt → upgrading-0.103.13-0.113.0.txt}

@@ -1,14 +1,33 @@
-# Upgrading 0.103.0 → 0.112.0 — stitchkit docs slice
+# Upgrading 0.103.13 → 0.113.0 — stitchkit docs slice
 
 > Generated from the `docs/` tree by `bun run gen:llms`. The index of all slices is
 > `llms.txt` in this package.
 
-Contains: Guide: Upgrading, 0.103.0 → 0.112.0  (docs/guide/upgrading.md)
+Contains: Guide: Upgrading, 0.103.13 → 0.113.0  (docs/guide/upgrading.md)
 
 ==============================================================================
-# Guide: Upgrading, 0.103.0 → 0.112.0  (docs/guide/upgrading.md)
+# Guide: Upgrading, 0.103.13 → 0.113.0  (docs/guide/upgrading.md)
 ==============================================================================
 
+## Released migration: 0.113.0
+
+**Who must act:** custom `AgentRuntimeStoreDriver` adapters and custom `AgentRuntimeStore` implementations of `seedConversationInput`. The built-in memory and SQLite stores already implement the contract.
+
+`seedConversationInput({ placement: 'after-settled' })` emits one `insert` history mutation per message. A driver's `history.apply` places the message ahead of the active row named `beforeMessageId`, or at the end when it is absent, and replaces a message already present in place:
+
+```ts
+// after: one more branch in history.apply
+if (mutation.type === 'insert') {
+  if (await exists(mutation.message.id)) return replaceInPlace(mutation.message)
+  if (mutation.beforeMessageId === undefined) return appendAtEnd(mutation.message)
+  const anchor = await activePosition(mutation.beforeMessageId) // refuse when missing
+  await shiftPositionsFrom(anchor) // every row at or after it moves one down
+  return insertAt(anchor, mutation.message)
+}
+```
+
+A custom `AgentRuntimeStore` that implements `seedConversationInput` itself places `'after-settled'` messages after the settled history and ahead of the first message of any active run, and keeps `'lead'` (or an absent `placement`) as before. Run `runAgentStoreConformance` against the adapter; `examples/agent-store-prisma/adapter.ts` shows the Prisma form.
+
 ## Released migration: 0.112.0
 
 **Who must act:** scripts using an array option together with bare positional arguments without a declared `positionals` policy, including calls to `parseCliArgs`.
@@ -876,32 +895,3 @@ mytool deploy --profile "${PROFILE:-production}" "$@"
 
 Dotted leaves such as `--filter.name` follow the same rule as their parent field. Declare a
 field as an array when repetition is meant.
-
-## Released migration: 0.103.0
-
-### Diagnostic journal startup preserves torn evidence and inspects archives
-
-**Who must act:** operators using `maxFiles: 1` who can encounter a torn active
-file, or whose retained files are unreadable by the writer. Clean single-file
-journals continue to work; successful startup now requires read access to every
-retained generation.
-
-```ts
-import { createDiagnosticJournal } from 'stitchkit/application/diagnostic-journal'
-
-// Before: a torn active file could be discarded during startup.
-await createDiagnosticJournal({ ...config, limits: { ...config.limits, maxFiles: 1 } })
-// After: retain the torn file while the fresh writer starts in a new active file.
-await createDiagnosticJournal({ ...config, limits: { ...config.limits, maxFiles: 2 } })
-```
-
-With single-file retention, a torn active file throws `DiagnosticJournalRecoveryError`
-from `stitchkit/application`, carrying `recovery` observations and releasing its lock
-without deleting evidence. Inspection I/O errors also fail startup and release the
-lock; restore filesystem access rather than treating that failure as a bad JSON row.
-Normal rotation retains the configured finite eviction policy.
-
-For archive inspection, replace ad hoc line parsing with `readDiagnosticJournal`
-from the same filesystem leaf. Handle its `frame | anomaly` stream explicitly;
-missing files, permissions and unsafe paths still throw. See the
-[journal guide](application-kernel.md#bounded-local-diagnostic-journal).

package/llms/{upgrading-0.75.0-0.102.0.txt → upgrading-0.75.0-0.103.0.txt}

@@ -1,14 +1,43 @@
-# Upgrading 0.75.0 → 0.102.0 — stitchkit docs slice
+# Upgrading 0.75.0 → 0.103.0 — stitchkit docs slice
 
 > Generated from the `docs/` tree by `bun run gen:llms`. The index of all slices is
 > `llms.txt` in this package.
 
-Contains: Guide: Upgrading, 0.75.0 → 0.102.0  (docs/guide/upgrading.md)
+Contains: Guide: Upgrading, 0.75.0 → 0.103.0  (docs/guide/upgrading.md)
 
 ==============================================================================
-# Guide: Upgrading, 0.75.0 → 0.102.0  (docs/guide/upgrading.md)
+# Guide: Upgrading, 0.75.0 → 0.103.0  (docs/guide/upgrading.md)
 ==============================================================================
 
+## Released migration: 0.103.0
+
+### Diagnostic journal startup preserves torn evidence and inspects archives
+
+**Who must act:** operators using `maxFiles: 1` who can encounter a torn active
+file, or whose retained files are unreadable by the writer. Clean single-file
+journals continue to work; successful startup now requires read access to every
+retained generation.
+
+```ts
+import { createDiagnosticJournal } from 'stitchkit/application/diagnostic-journal'
+
+// Before: a torn active file could be discarded during startup.
+await createDiagnosticJournal({ ...config, limits: { ...config.limits, maxFiles: 1 } })
+// After: retain the torn file while the fresh writer starts in a new active file.
+await createDiagnosticJournal({ ...config, limits: { ...config.limits, maxFiles: 2 } })
+```
+
+With single-file retention, a torn active file throws `DiagnosticJournalRecoveryError`
+from `stitchkit/application`, carrying `recovery` observations and releasing its lock
+without deleting evidence. Inspection I/O errors also fail startup and release the
+lock; restore filesystem access rather than treating that failure as a bad JSON row.
+Normal rotation retains the configured finite eviction policy.
+
+For archive inspection, replace ad hoc line parsing with `readDiagnosticJournal`
+from the same filesystem leaf. Handle its `frame | anomaly` stream explicitly;
+missing files, permissions and unsafe paths still throw. See the
+[journal guide](application-kernel.md#bounded-local-diagnostic-journal).
+
 ## Released migration: 0.102.0
 
 ### `view_file` receives an array, not an array encoded as text

package/llms/upgrading.txt

@@ -11,8 +11,8 @@ Contains: Guide: Upgrading, general  (docs/guide/upgrading.md)
 
 Load this slice, then every range slice between your installed version and the target:
 
-- 0.103.0 → 0.112.0: `llms/upgrading-0.103.0-0.112.0.txt`
-- 0.75.0 → 0.102.0: `llms/upgrading-0.75.0-0.102.0.txt`
+- 0.103.13 → 0.113.0: `llms/upgrading-0.103.13-0.113.0.txt`
+- 0.75.0 → 0.103.0: `llms/upgrading-0.75.0-0.103.0.txt`
 - 0.61.0 → 0.74.0: `llms/upgrading-0.61.0-0.74.0.txt`
 - 0.44.0 → 0.60.0: `llms/upgrading-0.44.0-0.60.0.txt`
 - 0.37.0 → 0.37.0: `llms/upgrading-0.37.0.txt`

package/llms-full.txt

@@ -77,12 +77,12 @@ own, recorded as an ADR.
 | `stitchkit/geo` | server (Bun or Node) | evolving | managed GeoIP reader generations, last-known-good reload and the optional MaxMind adapter |
 | `stitchkit/oauth` | browser **and** server | evolving | provider-neutral Authorization Code + PKCE request and one-shot callback transaction mechanics |
 | `stitchkit/google` | server (Bun or Node) | evolving | optional Google code exchange and verified OIDC identity adapter |
-| `stitchkit/observability` | server | stable<br>_redefined in 2 of the 57 minors since 0.56.2, most recently 0.92.0_ | request/tool event projections — `createObservability`, trace context, sanitisation |
+| `stitchkit/observability` | server | stable<br>_redefined in 2 of the 58 minors since 0.56.2, most recently 0.92.0_ | request/tool event projections — `createObservability`, trace context, sanitisation |
 | `stitchkit/testing` | tests on Bun or Node | stable | in-process generated clients over a real Fetch handler, the managed-resource conformance kit and manual time for schedules and retries |
 | `stitchkit/declaration` | browser + build and deployment tooling (Bun or Node) | evolving | `ProjectDeclarationSchema` — the one machine-readable statement a repository makes about itself |
 | `stitchkit/react` | browser + server rendering | stable | `createCursorQuery`, `createCacheBridge`, QueryClient and `ApiError` retry policy |
 | `stitchkit/react/keyboard` | browser | evolving | which part of a screen gets a key — ordered layers, one Escape per level, list zones with real focus; see [keyboard guide](keyboard.md) |
-| `stitchkit/agent-runtime` | server | evolving<br>_redefined in 23 of the 57 minors since 0.56.2, most recently 0.105.0_ | optional durable conversation/run loop, history, models, prompts, fencing and events |
+| `stitchkit/agent-runtime` | server | evolving<br>_redefined in 24 of the 58 minors since 0.56.2, most recently 0.113.0_ | optional durable conversation/run loop, history, models, prompts, fencing and events |
 | `stitchkit/agent-runtime/sandbox` | server | evolving | optional Linux Bubblewrap sessions, durable workspaces and a host HTTP credential gateway; see [sandbox guide](sandbox.md) |
 | `stitchkit/agent-runtime/testing` | tests on Bun or Node | evolving | credential-free replay, scripted provider faults, deterministic race controls and the agent-store conformance kit |
 | `stitchkit/agent-runtime/harness` | server | evolving | resource-aware process-local facade over the canonical Agent runtime; supervision stays outside |
@@ -94,7 +94,7 @@ own, recorded as an ADR.
 | `stitchkit/agent-runtime/sqlite/bun` | server (Bun) | evolving | durable built-in SQLite store for the agent runtime |
 | `stitchkit/agent-runtime/sqlite/node` | server (Node ≥ 22.5) | evolving | durable built-in SQLite store for the agent runtime |
 | `stitchkit-tui` | terminal (Bun) | evolving | optional official OpenTUI host over a caller-composed headless runtime |
-| `stitchkit/application` | browser + server | evolving<br>_redefined in 14 of the 57 minors since 0.56.2, most recently 0.110.0_ | managed resource graph, readiness, admission, schedules, subtree restart and bounded shutdown |
+| `stitchkit/application` | browser + server | evolving<br>_redefined in 14 of the 58 minors since 0.56.2, most recently 0.110.0_ | managed resource graph, readiness, admission, schedules, subtree restart and bounded shutdown |
 | `stitchkit/application/grammy` | server | evolving | a grammY bot as application resources — configuration, then polling with updates admitted by batch — and the webhook adapter |
 | `stitchkit/application/opentelemetry` | server | evolving | maps application snapshots onto an injected OpenTelemetry `Meter` |
 | `stitchkit/application/schemas` | browser + server | evolving | the application's snapshot, health and shutdown schemas alone, without the kernel |
@@ -5518,6 +5518,12 @@ await ticket.accepted
 const terminal = await ticket.result
 ```
 
+A stop policy's `when` receives the steps so far together with the same run context
+`prepareStep` gets — `context`, `run`, `signal`, `toolFenceLifecycle` and `contextUsage` —
+typed by the runtime's `CONTEXT`. A rule that depends on its own run (a receipt bound to the
+run's request, a per-tenant limit) reads it there; it needs no map of its own keyed by `runId`.
+The first policy to return `true` ends the run as `policy_stop` with its `name` as `policyName`.
+
 ## Bounded deferred tool catalogs
 
 Use `createDeferredAgentToolSurface` when a measured large catalog makes
@@ -6365,6 +6371,31 @@ implement the smaller `AgentRuntimeStoreDriver` rather than these eleven members
   calls this and nothing else, so an adapter that implements the interface has
   everything recovery needs
 
+### Writing history no run owns
+
+`seedConversationInput` writes messages no run owns, exactly once per `seedKey`; the key is
+recorded beside the history, so compaction or replacement of the messages never brings them
+back. It places them by `placement`:
+
+- absent or `'lead'` — committed user instructions ahead of the whole conversation;
+- `'after-settled'` — an exchange that already happened without a model (a confirmed voice
+  command and its receipt): committed `user` and completed `assistant` messages. They land after
+  the settled history and ahead of the first message of any run still active, so an active run's
+  input and answer stay adjacent and the next run reads the exchange before the input queued
+  behind it. With no active run they are the tail.
+
+```ts
+await store.seedConversationInput({
+  conversationId,
+  seedKey: `voice:${commandId}`,
+  placement: 'after-settled',
+  inputs: [ownerMessage, receiptMessage],
+})
+```
+
+The pair is ordinary history: it is in `loadSnapshot`, compaction may summarize it, and the next
+provider request carries it. No model is called and no run is created.
+
 ### Reading a conversation
 
 Two shapes of read, and the difference matters as a conversation grows.
@@ -10869,9 +10900,10 @@ One unconvertible schema does not take the connection down. The tool
 is skipped and **named** (`onSkippedTool`, or a stderr line by default), so a
 surface of two hundred tools is not lost to one.
 
-**A discovered schema is checked locally, except for its unions.** The arguments are
-coerced and parsed against the schema the server advertised, so a missing field or a
-wrong scalar is refused before anything is sent. A union in that schema is not judged
+**A discovered schema leaves unions and string formats to its server.** The arguments are
+coerced and parsed against the schema the server advertised. Reported structural failures
+(a missing field, wrong scalar type, unknown key or enum value) and bounds stay local.
+A union in that schema is not judged
 here: a server's presentation may join a discriminated union into one object whose
 shared field is an unlabelled `anyOf` (a stitchkit server with `flattenUnionInput`
 does), and judged against that, a wrong value could only be refused with every branch
@@ -10879,6 +10911,16 @@ listed. The value goes to the server, which holds the original schema, and its r
 names the one wrong field — `blocks.0.data.artifactId: expected string` rather than
 `No union branch matched — branch 1 … and 6 more branches`.
 
+String formats and patterns also lose the author's error messages in JSON Schema. A
+restored `date-time`, email, UUID or regex check would substitute Zod's generic wording.
+Those `invalid_format` refusals go to the remote instead: a bare `2026-10-12` reaches
+the original datetime schema, whose answer can explain that a full UTC timestamp such
+as `2026-10-12T00:00:00Z` is required. This also applies inside arrays, objects and
+record keys. A mixed refusal retains only its reported local reasons; a record-key
+wrapper stays local when it contains any of those reasons. A schema authored locally
+keeps its complete verdict, including custom errors and refinements. No second parse
+invents checks that the first parse did not reach. See [ADR 0272](../decisions/0272-a-restored-string-format-is-judged-by-its-server.md).
+
 MCP connection limits follow the operation phase. `initialize`, the initialized
 notification, `tools/list` and legacy endpoint discovery use `limits.discovery`;
 `tools/call` uses `limits.call`. A mounted call's handshake still uses discovery
@@ -16276,6 +16318,25 @@ analytics policy belong to the application.
 
 # Upgrading stitchkit
 
+## Released migration: 0.113.0
+
+**Who must act:** custom `AgentRuntimeStoreDriver` adapters and custom `AgentRuntimeStore` implementations of `seedConversationInput`. The built-in memory and SQLite stores already implement the contract.
+
+`seedConversationInput({ placement: 'after-settled' })` emits one `insert` history mutation per message. A driver's `history.apply` places the message ahead of the active row named `beforeMessageId`, or at the end when it is absent, and replaces a message already present in place:
+
+```ts
+// after: one more branch in history.apply
+if (mutation.type === 'insert') {
+  if (await exists(mutation.message.id)) return replaceInPlace(mutation.message)
+  if (mutation.beforeMessageId === undefined) return appendAtEnd(mutation.message)
+  const anchor = await activePosition(mutation.beforeMessageId) // refuse when missing
+  await shiftPositionsFrom(anchor) // every row at or after it moves one down
+  return insertAt(anchor, mutation.message)
+}
+```
+
+A custom `AgentRuntimeStore` that implements `seedConversationInput` itself places `'after-settled'` messages after the settled history and ahead of the first message of any active run, and keeps `'lead'` (or an absent `placement`) as before. Run `runAgentStoreConformance` against the adapter; `examples/agent-store-prisma/adapter.ts` shows the Prisma form.
+
 ## Released migration: 0.112.0
 
 **Who must act:** scripts using an array option together with bare positional arguments without a declared `positionals` policy, including calls to `parseCliArgs`.
@@ -22193,7 +22254,7 @@ Server-only optional application runtime. See the
 | `selectCompactableHistory` | function | which oldest whole complete turns may be summarised away — the half of compaction that needs no store (→ ADR 0142) |
 | `SelectCompactableHistoryOptions` / `CompactableHistory` | _type_ | message list, retained-turn count and evidence policy in; `leadingSummary`, `compactable` and `retained` out |
 | `createAgentSessionCoordinator` | function | strict process-local queue/interrupt/supersede lifecycle |
-| `AgentRuntimeStopPolicy` | _type_ | named custom AI SDK stop condition persisted and published on policy stop |
+| `AgentRuntimeStopPolicy` | _type_ | named stop rule persisted and published on policy stop; `when` receives the AI SDK `steps` plus the run context `prepareStep` gets (`context`, `run`, `signal`, `toolFenceLifecycle`, `contextUsage`) |
 | `AgentRuntimePrepareStep` | _type_ | per-run controlled step callback with typed domain context and managed run signal/fence |
 | `AgentContextOverflowError` | class | deliberate application budget refusal thrown from `loop.prepareStep`; terminalizes as `context_overflow` without classifying arbitrary error text |
 | `AgentRuntimeRecordIds` | _type_ | optional caller-provided input, run and assistant IDs for stable application records |
@@ -23053,7 +23114,7 @@ and approval path.
 | `ConnectionPhase` / `ConnectionOperation` / `ConnectionFailureContext` | _type_ | closed operation/phase vocabulary and measured read context for existing typed connection errors |
 | `ConnectionTimeoutError` / `ConnectionResponseTooLargeError` | class | bounded operation refusal, with phase, operation and finite observed bytes; safe tool projection retains the raw cause locally |
 | `defineOpenApiConnection` | function | declare one OpenAPI document (object, JSON text or JSON URL) as a tool surface |
-| `mountConnections` | function | discover every connection and return readonly `RuntimeToolDefinition`s for `mountAgent`; a call's arguments are parsed against the discovered schema before sending, except its unions, which the server judges (ADR 0266) |
+| `mountConnections` | function | discover every connection and return readonly `RuntimeToolDefinition`s for `mountAgent`; reported structural failures and bounds stay local, while unions and string-format/pattern refusals (including inside record keys) are judged by the server that owns the original schema (ADR 0266, ADR 0272) |
 | `ConnectionAuthorizationRequiredError` | class | typed reauthorization signal raised on `401` with credentials resolved again on the next call |
 | `ConnectionBudgetExceededError` | class | the mount would expose more foreign tools than its declared budget |
 | `ConnectionRequestError` | class | any non-`2xx` connection response other than the reauthorization signal |

package/llms.txt

@@ -17,15 +17,15 @@ Load the slice for each entrypoint your code imports — `import … from 'stitc
 - **stitchkit/tools/mcp**: [llms/tools-mcp.txt](llms/tools-mcp.txt) 1.8 KB
 - **stitchkit/tools/contract**: [llms/tools-contract.txt](llms/tools-contract.txt) 0.2 KB
 - **stitchkit/tools/invoker**: [llms/tools-invoker.txt](llms/tools-invoker.txt) 0.2 KB
-- **stitchkit/tools/connections**: [llms/tools-connections.txt](llms/tools-connections.txt) 3.8 KB
-- **stitchkit/cli**: [llms/cli.txt](llms/cli.txt) 45.3 KB → [llms/cli.2.txt](llms/cli.2.txt) 21.4 KB
+- **stitchkit/tools/connections**: [llms/tools-connections.txt](llms/tools-connections.txt) 3.9 KB
+- **stitchkit/cli**: [llms/cli.txt](llms/cli.txt) 45.3 KB → [llms/cli.2.txt](llms/cli.2.txt) 22.3 KB
 - **stitchkit/cli/publish**: [llms/cli-publish.txt](llms/cli-publish.txt) 1.5 KB
 - **stitchkit/remote**: [llms/remote.txt](llms/remote.txt) 1.0 KB
 - **stitchkit/contract**: [llms/contract.txt](llms/contract.txt) 31.9 KB
 - **stitchkit/primitives**: [llms/primitives.txt](llms/primitives.txt) 11.7 KB
 - **stitchkit/server**: [llms/server.txt](llms/server.txt) 44.2 KB → [llms/server.2.txt](llms/server.2.txt) 20.5 KB → [llms/server.3.txt](llms/server.3.txt) 45.6 KB → [llms/server.4.txt](llms/server.4.txt) 38.3 KB → [llms/server.5.txt](llms/server.5.txt) 22.9 KB
 - **stitchkit/observability**: [llms/observability.txt](llms/observability.txt) 46.9 KB
-- **stitchkit/agent-runtime**: [llms/agent-runtime.txt](llms/agent-runtime.txt) 42.7 KB → [llms/agent-runtime.2.txt](llms/agent-runtime.2.txt) 44.8 KB → [llms/agent-runtime.3.txt](llms/agent-runtime.3.txt) 42.1 KB
+- **stitchkit/agent-runtime**: [llms/agent-runtime.txt](llms/agent-runtime.txt) 43.2 KB → [llms/agent-runtime.2.txt](llms/agent-runtime.2.txt) 45.9 KB → [llms/agent-runtime.3.txt](llms/agent-runtime.3.txt) 42.2 KB
 - **stitchkit/agent-runtime/testing**: [llms/agent-runtime-testing.txt](llms/agent-runtime-testing.txt) 2.6 KB
 - **stitchkit/agent-runtime/harness**: [llms/agent-runtime-harness.txt](llms/agent-runtime-harness.txt) 11.0 KB
 - **stitchkit/agent-runtime/coding-tools**: [llms/agent-runtime-coding-tools.txt](llms/agent-runtime-coding-tools.txt) 5.0 KB
@@ -64,8 +64,8 @@ Load the slice for each entrypoint your code imports — `import … from 'stitc
 
 ## Upgrading slices
 - **Upgrading — how to move across versions**: [llms/upgrading.txt](llms/upgrading.txt) 12.7 KB
-- **Upgrading 0.103.0 → 0.112.0**: [llms/upgrading-0.103.0-0.112.0.txt](llms/upgrading-0.103.0-0.112.0.txt) 46.6 KB
-- **Upgrading 0.75.0 → 0.102.0**: [llms/upgrading-0.75.0-0.102.0.txt](llms/upgrading-0.75.0-0.102.0.txt) 44.8 KB
+- **Upgrading 0.103.13 → 0.113.0**: [llms/upgrading-0.103.13-0.113.0.txt](llms/upgrading-0.103.13-0.113.0.txt) 46.4 KB
+- **Upgrading 0.75.0 → 0.103.0**: [llms/upgrading-0.75.0-0.103.0.txt](llms/upgrading-0.75.0-0.103.0.txt) 46.3 KB
 - **Upgrading 0.61.0 → 0.74.0**: [llms/upgrading-0.61.0-0.74.0.txt](llms/upgrading-0.61.0-0.74.0.txt) 43.5 KB
 - **Upgrading 0.44.0 → 0.60.0**: [llms/upgrading-0.44.0-0.60.0.txt](llms/upgrading-0.44.0-0.60.0.txt) 43.9 KB
 - **Upgrading 0.37.0 → 0.37.0**: [llms/upgrading-0.37.0.txt](llms/upgrading-0.37.0.txt) 6.2 KB

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "stitchkit",
-  "version": "0.112.3",
+  "version": "0.113.1",
   "description": "Contract-first backend framework — one defineContract() into an HTTP API, MCP tools, AI-agent tools and a typed client. Bun and Node.",
   "keywords": [
     "bun",