Repository navigation
Conversation
mhegazy
force-pushed
the
mhegazy/sg-commit-findings-stderr
branch
from
October 9, 2026 16:03
a4debd2 to
8f2fb08
Compare
Commit and push findings went only into hookSpecificOutput.additionalContext, which Claude Code's asyncRewake path never reads: it shows Claude stderr, or stdout when stderr is empty. When the agentic reviewer's inner CLI printed a warning to stderr, Claude got that warning instead of the findings (anthropics/claude-code#99839, also anthropics/claude-code#92987). Findings now go to stderr for every event, as the Stop path already did. The inner CLI's stderr and the Agent SDK's own log lines go to the plugin's debug log.
mhegazy
force-pushed
the
mhegazy/sg-commit-findings-stderr
branch
from
October 9, 2026 16:06
8f2fb08 to
d3fed7d
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Human: @mhegazy
Summary
Commit and push findings could be replaced by an unrelated warning (anthropics/claude-code#99839, also anthropics/claude-code#92987). The findings went only into
hookSpecificOutput.additionalContext, which Claude Code's asyncRewake path does not read: it shows Claude the hook's stderr, or its stdout when stderr is empty. When the agentic reviewer's innerclaudeprinted a warning to stderr, Claude got the warning instead of the findings.What changes
No version bump here. A separate PR bumps the plugin to 2.0.13 and should merge after this one.
Test plan
New
tests/test_findings_delivery.py, 4 tests. Without the fix 3 of them fail. The plugin's suite passes, 192 tests, under pytest on Linux.End to end. Run in tmux with public Claude Code 2.1.292, the plugin loaded by
--plugin-dir, against a local fake API that scripts Claude and answers the plugin's review calls. The agentic reviewer's inner CLI is a stub that prints "claude.ai connectors are disabled" and exits 1, so the review falls back to a single call, which finds the bug. Claude writes a file with a shell-injection bug and commits it.Not verified: macOS and Windows.
Generated by Claude Code