Ralph Orchestrator is a framework that repeatedly runs AI-agent tasks until they finish or reach an iteration limit. Developers use it to coordinate autonomous coding work through command-line, web-dashboard, and MCP-server interfaces, with state managed per workspace. The catalogue entries provide agent skills, agents, instructions, and plugins for operating Ralph.
Borrowing it
Nothing to install: this file belongs to mikeyobrien/ralph-orchestrator. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mikeyobrien/ralph-orchestrator/main/.claude/skills/review-pr/SKILL.mdgit clone --depth 1 https://github.com/mikeyobrien/ralph-orchestratorWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mikeyobrien/ralph-orchestrator/review-pr)<a href="https://agentmods.dev/skills/mikeyobrien/ralph-orchestrator/review-pr"><img src="https://agentmods.dev/badge/skills/mikeyobrien/ralph-orchestrator/review-pr/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mikeyobrien/ralph-orchestrator/review-pr"><img src="https://agentmods.dev/badge/skills/mikeyobrien/ralph-orchestrator/review-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00032 | $0.00602 |
| Opus 5 | $0.00016 | $0.00301 |
| Sonnet 5 | $0.00006 | $0.00120 |
| Haiku 4.5 | $0.00003 | $0.00060 |
Grade A, and why
review-pr scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review PR
Run the ralph.reviewer.yml orchestration loop against a pull request. The loop checks out the PR in an isolated worktree, runs tests, reviews the diff, and produces a structured report.
Usage
/review-pr <PR number or URL>
Accepts: 207, #207, or https://github.com/.../pull/207
Execution
1. Parse the PR argument
Extract the PR number from the argument. Strip # prefix or extract from URL path.
2. Run the reviewer loop
ralph run -H ralph.reviewer.yml -p "Review PR #<N>"
Bash tool settings:
timeout: 600000(10 minutes)run_in_background: true
Use TaskOutput with block: true to wait for completion.
3. Display the report
Read and print .ralph/REVIEW-REPORT.md to the conversation.
If the report file doesn't exist (loop failed before the synthesizer hat), check for and display whatever intermediate files exist:
.ralph/review-scope.md— what was scoped.ralph/review-verification.md— test results.ralph/review-findings.md— review findings
4. Verify cleanup
Check that the review worktree was removed:
ls -d .worktrees/review-<N> 2>/dev/null
- If gone: cleanup succeeded, no action needed.
- If still present: warn the user but do NOT force-remove. Say what's there and let them decide.
Also note the presence of intermediate files (.ralph/review-scope.md, etc.) — they're useful for debugging but the user may want to clean them up later.
Error Handling
| Situation | Action |
|---|---|
| Ralph exits non-zero | Display error output. Suggest re-running with RALPH_DIAGNOSTICS=1 |
| Report file missing | Display intermediate files that do exist (scope, verification, findings) |
| PR argument missing | Ask the user for the PR number |
| PR argument unparseable | Ask the user to provide a bare number, #N, or full GitHub URL |
What This Skill Does NOT Do
- Does NOT validate the PR exists (the scoper hat handles that)
- Does NOT modify any source code (read-only review)
- Does NOT post comments to GitHub
- Does NOT own worktree cleanup (verifies only)
- Does NOT enable diagnostics by default
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 76 lines · 32 tokens per session scan A 5b92441c604c
review-pr is a skill published in the GitHub repository mikeyobrien/ralph-orchestrator (3,133 stars, last pushed yesterday), licensed MIT. It adds 32 tokens to every session and 602 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
code-review
Use when code has been written and needs validation before committing, or when the user asks for a code review or security check.
fast-review
Mandatory post-delegation gate: checks output completeness, verifies acceptance criteria compliance, flags regressions, produces PASS/FAIL verdict. Use when checking delegated work against acceptance criteria, running post-delegation gate, validating agent output before acceptance, verifying sub-agent completed its…
panel-majority-vote
Runs 3 isolated reviewer sub-agents; consolidates PASS/BLOCK verdict by majority. Use when user requests independent review of code changes, pull requests, design documents, or release notes.
project-consistency
Generates shared CSS variables, validates component naming conventions, creates layout pattern templates. Use when coordinating design system, theme, consistent styling, CSS variables, or component library across parallel agents.
review
Comprehensive code review workflow - parallel specialized reviews → synthesis.
refactor
Code refactoring workflow - analyze → plan → implement → review → validate.