chore: promote staging to staging-promote/1f209db0-23170138026 (2026-03-16 23:13 UTC)

[ironclaw-ci]

Auto-promotion from staging CI

Batch range: 63a23550d6b485de6eb3b9a8aefeee47de569ddd..c6128f4e41b5bd43d69a4432a6050df4d675590a
Promotion branch: staging-promote/c6128f4e-23170341776
Base: staging-promote/1f209db0-23170138026
Triggered by: Staging CI batch at 2026-03-16 23:13 UTC

Commits in this batch (17):

• a677b20 chore: promote staging to main (2026-03-10 15:19 UTC) (chore: promote staging to main (2026-03-10 15:19 UTC) #865)
• 6a1301b feat(i18n): Add internationalization support with Chinese and English translations (feat(i18n): Add internationalization support with Chinese and English translations #929) (chore: promote staging to main (2026-03-11 15:17 UTC) #950)
• 7e8c0fb chore: release v0.18.0 (chore: release v0.18.0 #885)
• 8391415 chore: update WASM artifact SHA256 checksums [skip ci] (chore: update WASM artifact checksums and version-pinned URLs #954)
• 8c2131d feat(embeddings): add EMBEDDING_BASE_URL for OpenAI-compatible embedding providers (feat(embeddings): add EMBEDDING_BASE_URL for OpenAI-compatible embedding providers #1082)
• ca6d9f6 fix(registry): bump versions for github, web-search, and discord extensions
• c4e098d Fix subagent monitor events being treated as user input (Fix subagent monitor events being treated as user input #1173)
• 877f117 feat(transcription): add Chat Completions API provider for audio transcription (feat(transcription): add Chat Completions API provider for audio transcription #1130)
• 0245c0f feat(orchestrator): read ORCHESTRATOR_PORT env var for configurable API port (feat(orchestrator): read ORCHESTRATOR_PORT env var for configurable API port #1113)
• de214c2 feat: add LLM_CHEAP_MODEL for generic smart routing across all backends (feat: add LLM_CHEAP_MODEL for generic smart routing across all backends #1081)
• 971b4c2 fix: web/CLI routine mutations do not refresh live event trigger cache (fix: web/CLI routine mutations do not refresh live event trigger cache #1255)
• 878a67c Refactor owner scope across channels and fix default routing fallback (Refactor owner scope across channels and fix default routing fallback #1151)
• fc18064 fix: resolve merge conflict fallout and missing config fields
• 026beb0 fix: cover staging CI all-features and routine batch regressions (fix: cover staging CI all-features and routine batch regressions #1256)
• 1f209db fix: bump channel registry versions for promotion (fix: bump channel registry versions for promotion #1264)
• ed0ed40 ci: isolate heavy integration tests (ci: isolate heavy integration tests #1266)
• c6128f4 fix: misleading UI message (fix: misleading UI message #1265)

Current commits in this promotion (2)

Current base: staging-promote/1f209db0-23170138026
Current head: staging-promote/c6128f4e-23170341776
Current range: origin/staging-promote/1f209db0-23170138026..origin/staging-promote/c6128f4e-23170341776

• ed0ed40 ci: isolate heavy integration tests (ci: isolate heavy integration tests #1266)
• c6128f4 fix: misleading UI message (fix: misleading UI message #1265)

Auto-updated by staging promotion metadata workflow

Waiting for gates:

• Tests: pending
• E2E: pending
• Claude Code review: pending (will post comments on this PR)

---

Auto-created by staging-ci workflow

[claude]

Code review

Found 6 issues:

1. [MEDIUM:75] Duplicated message construction logic in test vs production code

The approval context message is constructed in two places: production code at src/agent/thread_ops.rs:229-231 and again in the test helper extract_approval_message() at lines 2011-2015. The format string "Waiting for approval: {tool_name} — {desc_preview}. Use /interrupt to cancel." appears in both locations. This violates DRY and creates maintenance burden — if the message format changes, it must be updated in two places.

https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/src/agent/thread_ops.rs#L228-L235
https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/src/agent/thread_ops.rs#L2010-L2020

2. [MEDIUM:50] Semantic misuse of SubmissionResult enum

SubmissionResult::pending() (submission.rs:433-437) creates an Ok variant to avoid the "Error:" prefix in rendering. However, semantically this conflates success with a blocking/pending state. Per CLAUDE.md architecture guidance, prefer explicit variants over workarounds. Consider adding a dedicated Pending { message: String } variant to the enum that explicitly represents blocking states.

https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/src/agent/submission.rs#L430-L437

3. [MEDIUM:50] Unnecessary self.name clone in execute_on_start_with_state()

Line 1250 in src/channels/wasm/wrapper.rs clones self.name for the timeout error, but the function already cloned it into channel_name at line 1200 for use in the spawn_blocking error handler. Reuse the earlier clone to avoid the duplicate allocation.

https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/src/channels/wasm/wrapper.rs#L1243-L1253

4. [LOW:50] Unused shutdown_tx in start_message_stream_for_test()

The test helper at line 867 creates a shutdown_tx but never stores or returns it. This means the message stream can never be shut down. For test code this is acceptable (tests just drop the stream), but the pattern is worth noting for future refactoring.

https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/src/channels/wasm/wrapper.rs#L863-L880

5. [LOW:25] No explicit feature flag requirement on telegram integration test

test_private_messages_use_chat_id_as_thread_scope() in tests/telegram_auth_integration.rs (line 312) has #[cfg(feature = "integration")] guards but is not declared with required-features in Cargo.toml like e2e_thread_scheduling is. This creates inconsistency; adding required-features = ["integration"] to this test would be more explicit.

https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/tests/telegram_auth_integration.rs#L309-L330

6. [LOW:50] Potential race condition in E2E test (mitigated)

test_waiting_for_approval_message_no_error_prefix() in tests/e2e/scenarios/test_tool_approval.py (line 157) waits for message count to increase without explicit synchronization between frontend and backend approval state. However, the preceding await card.wait_for(state="visible", timeout=5000) provides reasonable mitigation. Low risk in practice.

https://github.com/anthropics/ironclaw/blob/fa736a1bd9f7ecf71ab37ca51550596f2fd2833f/tests/e2e/scenarios/test_tool_approval.py#L153-L180

Overall Assessment: Safe to merge. No critical or high-severity issues found. Medium issues are primarily DRY/design concerns rather than functional bugs.