ClaudeMap

lorenzini

Claude Code skills that wait for a third-party PR review and then adjudicate whether its verdict actually means pass — never inferring safety from silence.

Skills & Commandsskillscode-reviewcicoderabbitcopilot
Visit projectNanako0129/lorenzini

lorenzini is three reviewer-wait skills for Claude Code: coderabbit-review-wait (active on repositories with 10+ stars), copilot-review-wait (covers repositories under 10 stars, where CodeRabbit's OSS tier requires reviews to be triggered by hand — a vendor constraint, not a preference), and codex-review-wait (dormant since 2026-09-17, subscription paused). Each skill polls until a verdict lands on the current head commit and prints one machine-readable RESULT= line; they do not review code themselves, they adjudicate what a reviewer said about it. The governing rule is never to infer a pass from absence: a verdict requires an explicit positive completion marker, and no marker means keep polling. The repository's most valuable file may be docs/fail-open-ledger.md, which records five real bugs — all failing open, all reporting success wrongly — that produced the rules. Requires an authenticated gh and jq.

What it does

  • Three reviewer-wait skills — CodeRabbit (10+ star repos), Copilot (under 10), Codex (dormant since 2026-09-17) — picked by repo size, not taste
  • Polls until a verdict lands on the current head commit, then prints one machine-readable RESULT= line for the agent to branch on
  • The one rule: no explicit positive completion marker means keep polling — “I cannot see a problem” is never “there is no problem”
  • docs/fail-open-ledger.md documents five real fail-open bugs and the checks each one bought

When to use it

Agent-driven workflows where a bot review gates the next step: let Claude Code open a PR, wait for CodeRabbit or Copilot to actually finish on that head commit, and branch on an honest verdict instead of a timeout guess.

Getting started

git clone git@github.com:Nanako0129/lorenzini.git ~/side-project/lorenzini
for s in codex copilot coderabbit; do
  ln -s ~/side-project/lorenzini/$s-review-wait ~/.claude/skills/$s-review-wait
done

# requires gh (authenticated) and jq

Things to know

No LICENSE file — all rights reserved by default, so ask the author before redistributing. The codex-review-wait skill is dormant (subscription paused since 2026-09-17). The regression cases named in each SKILL.md are live public pull requests whose state can change between runs, so confirm behaviour changes with a controlled comparison before believing either result. The skills adjudicate reviewer verdicts; they do not review code themselves.

Repository

Nanako0129/lorenzini

Charted

Related on the map

  • guizang-product-video-skill

    Skill

    Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.

    • agent-skill
    • video
    • promo
    • marketing
    View

  • geo-sleuth

    Skill

    An agent skill that geolocates a photo with nothing readable in it — OSM geometry, elevation skylines and satellite tiles turn “where was this taken?” into a measured answer with an error radius.

    • agent-skill
    • geolocation
    • osint
    • openstreetmap
    View

  • Birdview

    Skill

    A skill that makes AI coding agents map the architecture and declare the blast radius before they touch a line of code — rendered as one standalone HTML page.

    • skill
    • architecture
    • code-visualization
    • change-scope
    View