
Review Fetch Phase
Fetches a GitHub pull request's review state and renders it into summary files for triage
- Repository
- prisma/orm
- Stars
- 47,663
- Licence
- Apache-2.0
- Files
- 11 files
What it does
Once installed, the agent can take a GitHub pull request URL and run a set of bundled Node.js scripts that download the PR's review threads, reviews and comments through the GitHub CLI (gh). It saves them as a canonical JSON file, checks that file's schema, then produces a Markdown report, a text summary and a list of unresolved review targets. It is only the data-gathering step of a larger review workflow and does not triage or change code.
Great for
- Maintainers of the Prisma repository or similar projects running a structured PR review workflow
- Developers who want a local, untracked snapshot of a PR's review discussion before triaging it
Skip it if
- Users without the GitHub CLI installed and authenticated, since fetching will fail without it
- Anyone wanting automatic code fixes or review decisions, because this skill only collects and renders data
- Sensitive or private repositories where review text should not be written to disk, since the artifacts contain full comment bodies and author logins
Try asking
- “Fetch the review state for https://github.com/OWNER/REPO/pull/123 and save it to wip/reviews”
- “Render a summary of the review threads for this PR without triaging them yet”
- “Refresh the review-state JSON and Markdown for PR 456 in a custom output folder”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 11 original files, the Apache-2.0 licence and an attribution note.
review-fetch-phase.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip review-fetch-phase.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip review-fetch-phase.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Run it from the skill directory so the relative script paths resolve.
- Check that the output folder is git-ignored before running; the guard script will stop generation otherwise.
- If gh reports TLS or certificate errors inside a sandbox, rerun outside the sandbox rather than disabling verification.
What's inside
- package.json
- scripts/extract-review-targets.mjs
- scripts/fetch-review-state.mjs
- scripts/guard-review-artifacts-ignored.mjs
- scripts/guard-review-artifacts-ignored.test.mjs
- scripts/render-review-state.mjs
- scripts/render-review-state.test.mjs
- scripts/review-artifacts.mjs
- scripts/summarize-review-state.mjs
- scripts/validate-review-state.mjs
- SKILL.md
Redistributed unmodified from prisma/orm under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.