← ClaudeAtlas

pr-walkthroughlisted

Generate an evidence-grounded markdown walkthrough for a pull request.
rp1-run/rp1 · ★ 38 · AI & Automation · score 75
Install: claude install-skill rp1-run/rp1
# PR Walkthrough Generate a slide-ready markdown walkthrough from direct pull request evidence. The output is a reviewer-orientation artifact with plain markdown fallback, not a review verdict or rendered slide reader. Use the pre-resolved `projectRoot`, `kbRoot`, `workRoot`, and `codeRoot` values from the generated Workflow Bootstrap section. Run all source-control commands from `codeRoot`, and write generated artifacts only under `workRoot`. Do not read `.rp1/work/pr-reviews/` or any existing `pr-review` artifacts as source material. The walkthrough source of truth is the target PR metadata, changed files, diffs, and commit summaries gathered in this workflow. ## STATE-MACHINE ```mermaid stateDiagram-v2 [*] --> collecting collecting --> publishing : evidence_ready publishing --> [*] : done ``` **On each phase transition**, report via: ```bash rp1 agent-tools emit \ --workflow pr-walkthrough \ --type status_change \ --run-id {RUN_ID} \ --step {CURRENT_STATE} \ --data '{"status": "running"}' ``` - `RUN_ID` comes from the generated Workflow Bootstrap section. - Derive `RUN_NAME` from the resolved target: use `"PR Walkthrough: PR #{number}"` when a PR number is available, otherwise `"PR Walkthrough: {branch_name}"`. - On the first emit only, include `--name "{RUN_NAME}"`. **Example sequence**: ```bash rp1 agent-tools emit --workflow pr-walkthrough --type status_change --run-id {RUN_ID} --name "PR Walkthrough: PR #42" --step collecting --data '{"sta