fix: enforce Markdown formatting in code review summary comments#139
Merged
fix: enforce Markdown formatting in code review summary comments#139
Conversation
The review prompt's summary template was treated as optional guidance, causing the model to flatten reviews into single-line pipe-delimited text. Fix by: - Adding a CRITICAL formatting constraint at the top of the prompt - Replacing vague placeholder approach with a concrete temp-file-based script that preserves multi-line Markdown - Adding a final checklist to verify formatting before finishing - Adding an anti-pattern entry for flattened summaries https://claude.ai/code/session_01HpnShWs5di3uhcDhkXncGr
| Automated Code Review. Commit: 4917787. Status: Approved. Findings: 0 Critical, 0 Important, 0 Suggestion. AGENTS.md Compliance: TypeScript strict mode N/A, no any types N/A, Biome linting N/A, all tests N/A, no secrets introduced. Summary: This PR fixes the code review summary being flattened into single-line text. The CRITICAL callout block, temp-file-based script pattern, and Final Checklist address the root cause. The removal of AGENTS.md examples and unused language rows keeps the prompt focused on this project. No issues found. CI Run: https://github.com/datenknoten/freundebuch/actions/runs/23204585884 |
- Fix broken nested code fences in Comment Format section (``` inside ```) by using ~~~ for the outer fence - Make AGENTS.md Compliance section dynamic: replace hardcoded generic placeholders with instruction to derive items from actual AGENTS.md - Remove ~55 lines of hypothetical AGENTS.md examples (Security-Focused, Performance-Critical, API Project) that added noise without value - Trim Language-Specific Focus table to TypeScript/JS and SQL only (the only languages in this project) - Remove duplicate PR_NUMBER and CI_RUN_URL definitions from Execution Context (already defined in Phase 1.1) - Make summary posting consistently file-based: use jq to build JSON payload from temp file for both create and update paths, avoiding shell escaping issues with $(cat ...) subshells https://claude.ai/code/session_01HpnShWs5di3uhcDhkXncGr
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code review passed. No issues found. This PR correctly addresses the root cause of flattened review summaries by adding an upfront CRITICAL formatting constraint, switching to a temp-file-based heredoc approach, and adding a Final Checklist. No correctness, security, or quality issues found.
|
🎉 This PR is included in version 2.67.6 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
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
2 participants
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.
The review prompt's summary template was treated as optional guidance,
causing the model to flatten reviews into single-line pipe-delimited
text. Fix by:
script that preserves multi-line Markdown
https://claude.ai/code/session_01HpnShWs5di3uhcDhkXncGr