crv
01Plugin Claude Code
Consensus code-review plugin for Claude Code (Opus + codex, optional pi advisor).
3 12d ago A
tokens not measured
original MIT
Plugin Claude Code
Two code-review strategies sharing one contract: /consensus-review (parallel multi-source panel with an Opus arbiter, codex consultant and strict sceptic) and /loop-review (sequential coverage-directed passes with evidence-backed cells and a CLEAN verdict enforced in code). Strictly read-only.
3 12d ago A
tokens not measured
original MIT
Agent
Review-only architect. Evaluates the ARCHITECTURAL IMPACT of a change (diff-scoped) — layers/slices, abstraction leakage, paradigms, complexity manageability. P0–P2. Not a line-by-line code review.
3 12d ago A 54 tokens
original MIT
Agent
Independent adversarial verifier of review findings. Fresh context, has NOT seen the synthesis. Tries to REFUTE each finding with evidence (code graph, repro, reading); confirms only what is concretely grounded. Read-only.
3 12d ago A 49 tokens
original MIT
Agent
Review-only review of the change itself and its IMPACT — correctness, regressions, security, migration/deploy safety, plus adjacent and dependent parts of the code. Verdict-first, P0–P2.
3 12d ago A 46 tokens
original MIT
Agent
Review-only check for code maintainability and quality — conventions, local maintenance cost, AI-slop, reuse/duplication, production-code contract alignment, scope control. P0–P2. Architectural boundaries and the complexity budget belong to arch-reviewer.
3 12d ago A 55 tokens
original MIT
Agent
Review-only agent for test quality — whether tests protect critical behavior and catch realistic regressions; mock/fixture drift, test layering, missing critical scenarios. P0–P2.
3 12d ago A 40 tokens
original MIT
arch-review
08Command
Architectural impact review of a change (diff-scoped) via the arch-reviewer agent.
3 12d ago A 18 tokens
original MIT
Command
One-shot architecture audit — layers, slices, contracts, paradigms, complexity.
3 12d ago A 15 tokens
original MIT
Command
Consensus code review (Opus arbiter + codex, optional pi advisor) across 4 dimensions (architecture/quality/impact/tests) for a PR or uncommitted changes; P0–P2 findings; strict sceptic by default; every flag also accepted with a leading -.
3 12d ago A 58 tokens
original MIT
doctor
11Command
Diagnose the consensus-review setup — consultants, code graph, contracts and live probes — and say what a broken piece would silently do to a review.
3 12d ago A 29 tokens
original MIT
Command
Review of a change and its impact on adjacent parts via the impact-reviewer agent.
3 12d ago A 16 tokens
original MIT
loop-review
13Command
Sequential, coverage-directed cyclic review of a PR or uncommitted changes — one immutable diff, differentiated discovery passes, per-cell evidence, adversarial verification, and a CLEAN verdict that cannot be produced by a failed run.
3 12d ago A 43 tokens
original MIT
Command
Maintainability and code-quality review via the quality-reviewer agent.
3 12d ago A 12 tokens
original MIT
test-review
15Command
Test-quality review via the test-reviewer agent.
3 12d ago A 9 tokens
original MIT