rigor: Skill for Claude Code

.claude/skills/rigor-reviewer/SKILL.md

rigor-reviewer is a skill for Claude Code from rigortype/rigor. It costs 62 tokens per session (1,177 once invoked), scanned A, original, MPL-2.0.

A review workflow for checking a Rigor draft pull request against the architect's acceptance conditions. A pull request is a proposed set of changes for review; this workflow gives an approval or needs-fix verdict but never merges it.

In plain words
What is it for?
Use it to review diffs, check acceptance items, list required fixes, and prepare approval material for a draft pull request.
Why use it?
It provides an independent check that the implementation matches the plan and catches problems before merge.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents.

This is rigortype/rigor's own configuration. It tells Claude Code how to work on rigor itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything rigor configures →

Reuse

Borrowing it

Nothing to install: this file belongs to rigortype/rigor. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/rigortype/rigor/master/.claude/skills/rigor-reviewer/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/rigortype/rigor

Made for: Claude Code.

Wrote 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.

agentmods badge for rigor-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/skills/rigortype/rigor/rigor-reviewer/github.svg)](https://agentmods.dev/skills/rigortype/rigor/rigor-reviewer)
Your own site
<a href="https://agentmods.dev/skills/rigortype/rigor/rigor-reviewer"><img src="https://agentmods.dev/badge/skills/rigortype/rigor/rigor-reviewer/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.

agentmods 80×15 button for rigor-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/skills/rigortype/rigor/rigor-reviewer"><img src="https://agentmods.dev/badge/skills/rigortype/rigor/rigor-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 62 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,177 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00062 $0.01177
Opus 5.5 $0.00025 $0.00471
Sonnet 5.5 $0.00012 $0.00235
Haiku 4.5 $0.00006 $0.00118

Measured yesterday against content hash 28810c9c44a8, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

rigor-reviewer 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 yesterday.

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.

.claude/skills/rigor-reviewer/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review one pull request

You are the reviewer: a fresh agent that did not write the change. The coordinator lands the PR and routes each fix back to the lane that wrote it (docs/agents/parallel-lanes.md). The brief gives you the base commit, the PR and its issue, and any edge-case table the coordinator wrote; this skill gives you everything else. A brief that contradicts this skill wins for that review, except on the Read-only bullet under Set up, which no brief loosens.

Set up

  • Read-only. Make no edits, pushes or comments on the PR. Every command goes through nix develop --command, and rspec and rigor get stdin from /dev/null. references/ is read-only. Never run make verify, make check or the whole suite. Fixtures and scratch files go under the review worktree's tmp/. No git stash, no git submodule deinit, no pkill/killall; stop only processes you started. A survey target gets a private copy (docs/agents/measurement.md), and you run one heavy job (a corpus rigor check) at a time.

  • Worktree. Review in a detached worktree at the PR's head SHA with its bundle cloned:

    git worktree add --detach <path> <sha>
    # from the main clone:
    mkdir -p <path>/vendor <path>/.bundle
    cp -Rc vendor/bundle <path>/vendor/bundle   # macOS clone; elsewhere cp -R --reflink=auto
    cp .bundle/config <path>/.bundle/
    

    Remove it with git worktree remove --force <path> when you report.

Review

  • Probe, not re-run. For a change that makes Rigor answer something, build the wrong-answer shapes as fixtures. Run that tree's bundle exec exe/rigor check --no-cache on them with a same-file positive control, and compare against what Ruby does. Read docs/agents/measurement.md first, and re-run the brief's edge-case table when it has one.
  • Run the precision half too. When the change touches how the engine records ADR-46 dependencies, run every spec/rigor/analysis/*incremental*_spec.rb that asserts a reader is NOT re-checked (not_to include on affected), not only the new fixture's file. #2136 passed two reviews and failed CI twice on those; verify-changed runs them for such an edit.
  • Prove the specs. Remove each part of the fix and report which examples fail. An example that survives every removal is a finding.
  • Name the class, not one instance. When a finding is one spelling of a wider shape (every self-changing block, every reader of one table), say so and list each instance you found, so the lane fixes the class in one round.
  • Severity is defined in docs/agents/contribution-flow.md § "Landing a pull request". Mark each finding introduced or pre-existing, and reproduce a pre-existing one on the base commit. A pre-existing finding is filed, not fixed in the PR, unless the change makes it reachable more often.
  • Delta rounds come back to you by message. Re-check every earlier finding with its original repro, then hunt for regressions the fix introduced, around the fix and in the code next to it.

Read the full file on GitHub · 86 lines

Changes

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.

  1. yesterday Changed · +18 lines · +12 tokens per session 28810c9c44a8
  2. 11d ago First seen · 68 lines · 50 tokens per session scan A e4303299d162

Subscribe to this mod's changes

rigor-reviewer is a skill published in the GitHub repository rigortype/rigor (107 stars, last pushed today), licensed MPL-2.0. It adds 62 tokens to every session and 1,177 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-26.

Related

rigor-reviewer compared with skills from other repositories

rigor-reviewer next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
rigor-reviewerrigortype/rigorSkill107todayMPL-2.0A62Claude Code
autoreviewopenclaw/clawhubSkill9.5kyesterdayMITA17Claude Code, Codex
agent-teams-simplify-and-hardenLeoYeAI/openclaw-master-skillsSkill2.2k5d agoMITA115Claude Code, Codex
omh-code-reviewrlaope/oh-my-hermesSkill5.2ktodayMITA48Claude Code, Codex
revdiff-planumputun/revdiffSkill945yesterdayMITA84Claude Code
explore-unknownsdzhng/skillsSkill1.4k2d agoMITA162Claude Code, Codex
re0-mergeLilMGenius/paperthinSkill1.4k4d agoMITA70Claude Code