loki-mode 11.72.3 → 11.72.5

package/SKILL.md

@@ -3,7 +3,7 @@ name: loki-mode
 description: Runs an autonomous software factory that knows what it is supposed to deliver and proves it did, with a signed receipt. Use when the user says Loki Mode or asks to build, fix or verify software autonomously.
 ---
 
-# Loki Mode v11.72.3
+# Loki Mode v11.72.5
 
 **Current entry points (Loki 10 engine):** `loki "<task>"`, `loki owner/repo#N` and `loki quick "<task>"` run the Loki 10 engine and end with a signed receipt (`loki verify`). `loki backlog owner/repo --all` runs many issues in parallel. Guide: `docs/v10/GUIDE.md`. The skill instructions below, and `loki start`, describe the previous bash loop. PRD-file starts (`loki start ./prd.md`) and the opencode provider still run on it until FC38-SWEEP ports them; the legacy-engine command and the LOKI_ENGINE switch remain removed. Prefer `loki owner/repo#N` for issues.
 
@@ -476,4 +476,4 @@ Implemented, planned and deprecated items, with the version each landed in: `ski
 
 ---
 
-**v11.72.3 | [Autonomi](https://www.autonomi.dev/) flagship product | core skill**
+**v11.72.5 | [Autonomi](https://www.autonomi.dev/) flagship product | core skill**

package/VERSION

@@ -1 +1 @@
-11.72.3
+11.72.5

package/dashboard/__init__.py

@@ -7,6 +7,6 @@ Modules:
     control: Session control helpers (status, atomic writes, origin check)
 """
 
-__version__ = "11.72.3"
+__version__ = "11.72.5"
 
 __all__ = ["__version__"]

package/docs/CLI-REFERENCE.md

@@ -1,6 +1,6 @@
 # CLI reference
 
-Generated from `loki-ts/src/cli/registry.ts` by `scripts/generate-stale-zero.sh` for v11.72.3. Do not edit by hand.
+Generated from `loki-ts/src/cli/registry.ts` by `scripts/generate-stale-zero.sh` for v11.72.5. Do not edit by hand.
 Hidden and legacy-only commands are omitted; see `docs/v10/CLI-MODERN.md` for the full inventory.
 
 | Command | Description |

package/docs/INSTALLATION.md

@@ -2,7 +2,7 @@
 
 Loki Mode, by [Autonomi](https://www.autonomi.dev/), turns a spec or an issue into a verified pull request with a receipt you can check. This page covers installing it, a first run, and the options around it.
 
-**Version:** v11.72.3
+**Version:** v11.72.5
 
 **Engine note:** `loki start` is the primary command. `loki start`, `loki "<task>"`, `loki owner/repo#N` and `loki quick` all run the Loki 10 engine, whatever the flag order or input form (PRD file, issue ref, task or one-word brief). The previous loop runs only on `--engine legacy`, `LOKI_ENGINE=legacy`, or for one call when a flag Loki 10 cannot honor is present. See the [user guide](v10/GUIDE.md) and the [5-minute quickstart](QUICKSTART.md).
 

package/docs/dev/architecture-reference.md

@@ -84,6 +84,32 @@ Observability: SDK failures emit a structured `capability_degraded` record to
 `.loki/events.jsonl`; `.loki/app-runner/first-preview.json` records
 time-to-first-preview write-once (bash route only).
 
+## Anthropic SDK upgrade capability matrix (RQ-104, 2026-10-23)
+
+Pins (exact, all sites): `@anthropic-ai/sdk` 0.124.0 -> 1.2.0 (loki-ts/package.json, loki-ts/bun.lock, docker/Dockerfile peer pin); `@anthropic-ai/claude-agent-sdk` 0.3.293 -> 0.3.309 (loki-ts/package.json, package.json, package-lock.json, bun.lock, docker/Dockerfile). `tests/test-sdk-version-sync.sh` asserts both. Sources: sdk CHANGELOG 0.125.0..1.2.0, agent-sdk CHANGELOG 0.3.294..0.3.309.
+
+| Change | Status |
+|---|---|
+| sdk 1.0.0 BREAKING "new API version 2026-10-21" | WIRED at compile level only: loki-ts/src/runner/sdk_invoker.ts:19,311 (messages.create, output_config) compiles unchanged under tsc rc=0 and sdk_invoker tests pass. Live behavior at API version 2026-10-21 is NOT PROVEN (no API key, no paid run) |
+| sdk 0.129.0 `claude-sonnet-5-5`, 0.128.0 `claude-opus-5-5`, 0.132.0 `claude-haiku-5-5` model ids | WIRED: model ids are passed through as strings by sdk_invoker.ts; no allowlist to change |
+| sdk 0.135.0 `speed` param and `usage.speed` | NOT ADOPTED: the cost ledger (util/child_ledger.ts:54) has no fast-mode price table; adopting speed without pricing would misreport cost |
+| sdk 0.134.0 native OpenTelemetry tracing (auto-on with a global tracer provider) | NOT ADOPTED: Loki registers no global tracer provider, so the SDK stays inert; no opt-in needed |
+| sdk 0.126.0 compaction param, 0.127.0 `compactBeforeNextTurn()`, 0.129.0 tool-call-while-streaming, tool runner fixes (0.128.0, 1.2.0) | NOT ADOPTED: sdk_invoker is a one-shot messages.create judge, not a tool runner or multi-turn loop |
+| sdk 0.129.0 cache diagnostics on Message | NOT ADOPTED: Loki already accounts cache tokens from usage (util/child_ledger.ts cache_creation_tokens, cache_read_tokens; cost_breakdown); diagnostics would add a second source of truth |
+| sdk 0.129.0 `between_tools` thinking | NOT ADOPTED: judge path clamps thinking by effort (sdk_invoker.ts clampThinkingForEffort); between_tools is not offered to judges |
+| sdk 0.135.0/0.136.0 streaming fixes (unread stream close, unknown SSE events) | NOT ADOPTED: sdk_invoker does not stream |
+| sdk 0.130.0-1.2.0 Admin API (analytics, spend limits, RBAC, plugins, webhooks, message threads, usage/cost reports), Managed Agents additions | NOT ADOPTED: no Loki feature calls these APIs |
+| agent-sdk 0.3.302 refused structured output now an error with no payload | NOT AFFECTED: quality_gates.ts:1616 reads `e["structured_output"]` and fails closed to null when it is absent or not an object (falls back to parsing `result`); a refusal with no payload therefore yields null, the same outcome as before. Refusals are also handled by sdk_stream_parser.ts refusal frames |
+| agent-sdk 0.3.300 `thinking: disabled` runs adaptive on models that cannot disable | NOT ADOPTED: providers.ts:846 query() options pass no thinking setting |
+| agent-sdk 0.3.302 fork subagents on by default in SDK sessions | NOT ADOPTED: accepted default, effect on Loki runs unmeasured; `CLAUDE_CODE_FORK_SUBAGENT=0` is the opt-out |
+| agent-sdk 0.3.302 omitted `permissionMode` defaults to `default` on third-party providers | N/A: Loki always passes an explicit permissionMode (providers/claude_flags.ts:1052) |
+| agent-sdk 0.3.301 full-disk login refresh error is `invalid_request`, not `authentication_failed` | NOT ADOPTED: sdk_stream_parser.ts:278 keys only on `authentication_failed`; a full-disk refresh failure surfaces as a generic error, accepted until observed |
+| agent-sdk 0.3.300 PreToolUse/PermissionRequest `updatedInput` validated, unknown keys dropped | N/A: runSdkHostGuard (providers.ts:784) returns no updatedInput |
+| agent-sdk 0.3.296 `claude_code_version` in the initialize control response | NOT ADOPTED: SDK_BUNDLED_CLAUDE_CODE (providers.ts:364) is read before any session starts and is guarded by a test against the installed SDK claudeCodeVersion; the initialize value only exists after spawn |
+| agent-sdk 0.3.295 SDK passes option values as `--flag=value` | N/A: Loki consumes the SDK options API, not CLI argv, on the SDK route |
+| agent-sdk 0.3.301 `BUILTIN_TOOL_NAMES`, sdk MCP `permission_policy`; 0.3.304 `suspendSseStream` (browser); 0.3.296 `autoCompactWindow`; 0.3.307 `wakeup_source`; 0.3.305 `startup()` timeout fix | NOT ADOPTED: no existing Loki feature maps to these options |
+| agent-sdk 0.3.309 `extractFromBunfs()` per-user temp dir | NOT ADOPTED: Loki does not run the SDK from a Bun-compiled binary |
+
 ## Phase 1 / RARV-C Closure Env Vars
 
 Default-on in the Bun runner (see `CHANGELOG.md` v7.x entries; documented in `skills/quality-gates.md:88-110`). Set to `0` to disable; set to `1` to force-enable on the bash route.

package/docs/v10/FAILURE-CLASSES.md

@@ -1739,6 +1739,7 @@ L0 review rule (ENGINE-LAWS.md L0): a fix that adds an `if` or a regex about the
 - Siblings swept: every test under tests/engine10 that reads src repo-wide was listed; the three failing here are the only ones red on main. The same weakness exists for any future slice, so the sweep rule is that a slice touching loki-ts/src runs budget.test.ts and l0_guard.test.ts, not only its own files.
 - Mechanism: untouchedUntracked moved to features/untouched_untracked.ts; the producer fingerprint code moved from engine10/cache.ts to util/producer_fingerprint.ts; the implement lead now says to run tests with the exact command the Project Model lists instead of naming a runner. No cap was raised and no allowlist entry was added.
 - Fixture: the existing guards themselves, red before (1513, a pytest literal in context.ts, 5025) and green after. Mutants: five blank lines appended to commit_filter.ts turn the e10ext cap red, five to cache.ts turn the core cap red, and a pytest literal put back into context.ts turns the L0 guard red.
+- Recurrence (pd/mainred): the same slice gate weakness left two MORE engine10 guards red on origin/main. (1) never_below_raw "L1 static guard" (3 tests): 9efb8eca0 changed the seal.ts receipt field to `model: externalRun(o) ? EXTERNAL_MODEL_LABEL : ctx.model`, a record of the run model that configures no session; the guard judged it an unvetted pin. (2) the implement.test.ts brief golden (len 2781): 85979c3af made the FIXED_RULES lead stack-neutral (len 2791, sha 55cb50f6...) without re-capturing it. Product right in both; guard allowlist and golden were stale. Fix: allowlist the seal.ts receipt line in the guard, re-capture the golden. Mutants: reverting the seal.ts allowlist line turns 3 L1 tests red; changing one byte of FIXED_RULES turns the golden red.
 
 ## FC-FULL-SUITE-IGNORES-PROJECT-INTERPRETER The deep-verify full pytest suite ran under bare `python`, not the project's own interpreter, so a repo with a working .venv and no `python` on PATH could never prove its suite
 - User-visible vs raw: user-visible. An uninterrupted run on a Python fixture with a working .venv and ruff ended PARTIAL because the full-suite check could not run (bare `python` missing or without pytest), while per-test runs under the same repo used the venv.