From review-crucible
This skill should be used when the user asks to "review this PR", "grill this PR", "deep code review", "pre-merge review", "find bugs in this diff", "check this branch before merge", "check tests", "check error handling", "check comments", "check types", "check project rules", "check prior PR comments", "simplify this change", or wants a strict high-signal review of pull requests, diffs, branches, or recent changes covering project rules, bugs, tests, errors, security boundaries, types, comments, prior PR discussion, design challenge, and simplification. Use this when correctness matters more than speed and findings must be validated before reporting. Do not use for one-off style nits, single-file readability passes unrelated to recently changed code, or generation of new features.
How this skill is triggered — by the user, by Claude, or both
Slash command
/review-crucible:review-crucibleThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Review Crucible is a source-neutral review workflow for Claude Code, Codex, and
Review Crucible is a source-neutral review workflow for Claude Code, Codex, and
Amp. Its behavior is defined by the canonical references in references/.
Load only the references needed for the task.
references/model-policy.md so review passes use the strongest verified host
model and the highest supported effort.references/review-orchestration.md: scope discovery, pass selection,
targeted activation, host routing, and validation ownership.references/reviewer-roles.md: canonical behavior for each reviewer role.references/review-rubric.md: severity, confidence, false-positive filters,
validation gate, and output format.references/agent-roster.md: role names and host check/agent mapping.references/model-policy.md: strongest-model and highest-effort defaults.references/compatibility.md: Claude Code, Codex, and Amp limits.references/github-commenting.md: inline PR comment discipline.references/rust-best-practices-integration.md: Rust handoff rules.reviewer-roles.md.AGENTS.md, CLAUDE.md, contribution, testing, release, and
language/framework instruction files.review-orchestration.md.finding-validator or parent/orchestrator logic
when the host cannot run a validator subagent.review-rubric.md.github-commenting.md.reviewer-roles.md#code-simplifier.agents/*.md can launch focused reviewers.
The generated agents preload this skill so references are available..codex/agents/*.toml custom agents for
reviewer fan-out. Codex only spawns subagents when explicitly asked..agents/checks/*.md and the Amp AGENTS.md snippet. Amp checks
run as separate review subagents, but the parent thread owns final
deduplication, validation, fixes, and comment posting.If a host cannot access a reference at runtime, say which reference was unavailable and continue only within the documented host limit.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub tailrocks/review-crucible