fix: preserve provider tool-stop failures

[seonghobae]

Summary

• preserve provider-originated 409 tool_execution_stopped through chat and raw passthrough retries
• map provider terminal tool state to the existing ToolFallbackStoppedError fail-closed contract
• preserve the same contract when a streaming provider response fails before SSE content starts
• make the tool fallback test module execute its full pytest collection when run directly

Validation

• uv run --frozen --with pytest --with ruff python -m pytest -q tests/test_provider_reliability.py tests/test_tool_execution_fallback.py (115 passed)
• uv run --frozen --with pytest --with ruff python -m compileall -q contextual_orchestrator tests
• git diff --check

This PR is intentionally stacked on upstream PR #771 and must not be merged independently of that parent.

---

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: c2a787b5-54a6-4201-b236-e9b0ceb2ce6f

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[seonghobae]

Maintainer exact-head audit for PR 811

Remote exact head: f0b0dd5
Base: cc806cd
Remote state: OPEN, non-Draft, MERGEABLE, BLOCKED; review decision REVIEW_REQUIRED and no formal approval.

The current diff is bounded to the provider tool-stop transport boundary and its regression tests. No unresolved review comments are present on this exact head. Local focused verification on the exact head passed 115 tests:
uv run --with pytest pytest -q tests/test_provider_reliability.py tests/test_tool_execution_fallback.py

Hosted checks currently have 8 skipped/completed runs and 9 queued runs, including the required workflow bootstrap, Strix, Scorecard, dependency review, OSV, Semgrep, Trivy FS, scheduler sweep, and close-empty. Queued runs are not success evidence. The PR is a dependent stack item over PR 771, so mergeability and base identity must be rechecked after the parent changes.

Decision: WAIT_AND_REMEDIATE. No merge, bypass, or approval was performed.

[seonghobae]

Exact-head stacked reconciliation completed.

• Parent PR fix: classify tool-runtime failures and fail safely #771 HEAD: 276ed4f04a065735577ab81e065dad54250b90b3
• PR fix: preserve provider tool-stop failures #811 HEAD: 809b07a1992bc442e36c9312fe055da2d6a94bd0
• The stale parent conflict was resolved through a normal merge in the fork branch; no force push or protected bypass was used.
• The resulting tree preserves provider-originated HTTP 409 tool_execution_stopped mapping, no provider URL/body leakage, no retry/failover after an ambiguous tool result, streaming SSE fail-closed output, and the parent provider response boundary.
• Focused provider/tool-fallback suite: 116 passed.
• Full exact-tree suite: 1540 passed in 537.30s.
• compileall and git diff --check: passed.

The PR remains dependent on #771. Hosted required Checks and an independent formal approval remain authoritative; no merge or approval claim is made while they are pending.

[devin-ai-integration]

Devin Review found 1 new potential issue.

[devin-ai-integration]

📝 Info: Streaming error body consumed before re-raise

For a non-tool-stop HTTPError, _is_tool_execution_stopped(exc) reads the error body via exc.read(65536) before raise re-raises it, so any later reader of the body gets an empty stream.

---

Was this helpful? React with 👍 or 👎 to provide feedback.