Implement Claude PR BugScan workflow (Draft→Ready + push re-scan)#9
Open
Implement Claude PR BugScan workflow (Draft→Ready + push re-scan)#9
Conversation
Strips the Python project template cruft and replaces it with a focused GitHub Actions setup for automated Claude PR code review: - .github/workflows/claude_pr_review.yml: reusable workflow_call that checks out both the target repo and this workflow repo, loads the Codex Code Review prompt, and runs anthropics/claude-code-action@v1 - .github/workflows/ai_pr_review.yml: caller workflow triggered on ready_for_review / synchronize / reopened, with draft guard and concurrency cancel-in-progress - prompts/codex_code_review_prompt.md: verbatim official Codex Code Review prompt from developers.openai.com cookbook - README.md: usage instructions, trigger table, versioning guidance - .gitignore: trimmed to repo-relevant patterns only https://claude.ai/code/session_018moumKtzZQiSiyGxVUFGdS
Add explicit "one-time tag setup" step, explain that target repos only need the two-line caller workflow (everything else runs in this repo), and note that updates here propagate automatically on next tag release. https://claude.ai/code/session_018moumKtzZQiSiyGxVUFGdS
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b6ceb2c1c4
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| - name: Checkout workflow repo (for prompt) | ||
| uses: actions/checkout@v4 | ||
| with: | ||
| repository: safurrier/python-collab-template |
There was a problem hiding this comment.
This checkout pulls safurrier/python-collab-template from the default branch because no ref is set, so callers pinned to @v1 still read whatever prompt is currently on main. That makes the reusable workflow non-deterministic and can break existing consumers if prompts/codex_code_review_prompt.md is changed or moved later.
Useful? React with 👍 / 👎.
| run: | | ||
| if [ -n "${{ inputs.prompt_override }}" ]; then | ||
| echo "text<<EOF" >> "$GITHUB_OUTPUT" | ||
| printf "%s\n" "${{ inputs.prompt_override }}" >> "$GITHUB_OUTPUT" |
There was a problem hiding this comment.
The step injects ${{ inputs.prompt_override }} directly into a quoted shell string, which can break parsing when the override contains double quotes or other shell-significant characters (a common case for prompt text). In that scenario the workflow fails before Claude runs, so prompt customization advertised in the README is unreliable.
Useful? React with 👍 / 👎.
Context files agent: - prompts/context_files_prompt.md — verbatim prompt for generating / updating AGENTS.md and CLAUDE.md; uses $ARGUMENTS substitution for mode (auto/quick-start/update/evaluate) and scope - .github/workflows/context_files_agent.yml — reusable workflow_call; accepts agent_args (mode+scope) and prompt_override; needs contents: write to commit context files back to PR branches - .github/workflows/context_files_pr.yml — caller workflow triggered on pull_request opened + ready_for_review MkDocs documentation site: - mkdocs.yml — Material theme, nav covering all pages - .github/workflows/docs.yml — build-check on PRs, gh-deploy on main - docs/index.md — overview and architecture diagram - docs/user-stories.md — solo dev, team lead, DevEx engineer stories - docs/using.md — step-by-step setup, trigger table, prompt override guide - docs/contributing.md — three-file pattern for adding a new agent - docs/agents/index.md — agent catalog and shared design principles - docs/agents/pr-review.md — review agent triggers, caller snippet, prompt - docs/agents/context-files.md — context agent modes, agent_args, permissions Repo context files (generated via context files prompt logic): - AGENTS.md — WHY/WHAT/HOW, repo map, validated commands, gotchas - CLAUDE.md — symlink to AGENTS.md (Claude Code auto-discovery) - .gitignore — remove CLAUDE.md exclusion so symlink is tracked https://claude.ai/code/session_018moumKtzZQiSiyGxVUFGdS
Replaces per-agent reusable workflows and flat prompts/ files with a single generic skill runner and structured SKILL.md files following the Claude Code skill convention. Skills (replaces prompts/): - skills/codex-code-review/SKILL.md — frontmatter + verbatim Codex body - skills/context-files/SKILL.md — frontmatter + full context files prompt Generic workflow (replaces claude_pr_review.yml + context_files_agent.yml): - .github/workflows/claude_pr_agent.yml — accepts skill: (name or path), args: ($ARGUMENTS substitution), prompt: (raw bypass); resolves bare names from central repo skills/ and paths from target repo checkout; strips YAML frontmatter via awk before sending body to Claude Skill resolution: "codex-code-review" → skills/codex-code-review/SKILL.md (central) ".claude/skills/my-skill" → .claude/skills/my-skill/SKILL.md (target repo) "./skills/my-skill" → skills/my-skill/SKILL.md (target repo) "./skills/my-skill.md" → flat file, no directory (target repo) Sequencing via needs:: caller workflows chain skills as separate jobs; each job calls claude_pr_agent.yml with a single skill; needs: enforces ordering without requiring dynamic steps. Updated callers: - ai_pr_review.yml → skill: "codex-code-review" - context_files_pr.yml → skill: "context-files", args: "auto ." Updated docs: - docs/agents/index.md — skills pattern and shared workflow explanation - docs/agents/pr-review.md — skill frontmatter, local skill path override, sequencing example with needs: - docs/agents/context-files.md — args: input, local skill path override - docs/contributing.md — two-file pattern (SKILL.md + doc page, no new workflow) - docs/using.md — skill: input, local skill paths, mixed central+local example https://claude.ai/code/session_018moumKtzZQiSiyGxVUFGdS
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.
Strips the Python project template cruft and replaces it with a focused
GitHub Actions setup for automated Claude PR code review:
checks out both the target repo and this workflow repo, loads the
Codex Code Review prompt, and runs anthropics/claude-code-action@v1
ready_for_review / synchronize / reopened, with draft guard and
concurrency cancel-in-progress
Review prompt from developers.openai.com cookbook
https://claude.ai/code/session_018moumKtzZQiSiyGxVUFGdS