Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/plumpslabs/matchaWrote 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/commands/plumpslabs/matcha/matcha-review)<a href="https://agentmods.dev/commands/plumpslabs/matcha/matcha-review"><img src="https://agentmods.dev/badge/commands/plumpslabs/matcha/matcha-review/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/commands/plumpslabs/matcha/matcha-review"><img src="https://agentmods.dev/badge/commands/plumpslabs/matcha/matcha-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00020 | $0.01734 |
| Opus 5.5 | $0.00008 | $0.00694 |
| Sonnet 5.5 | $0.00004 | $0.00347 |
| Haiku 4.5 | $0.00002 | $0.00173 |
Grade A, and why
matcha:review 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 9d 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 — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/matcha:review
The gate between code and "done". Nothing ships until this passes.
Gate Execution (risk-tiered)
- L0/L1 — async/non-blocking. Agent continues with the next low-risk step while the review runs; the verdict lands in background (
.agents/reports/reviewer-<YYYY-MM>.md). A 🔴 finding escalates to blocking. - L2/L3 — blocking. Agent stops until verdict PASS (L2) or EXPERT_REQUIRED sign-off (L3).
Always run matcha_review_validate on the rendered verdict before finalizing — malformed verdicts are caught at write time, not later.
Fallback (reviewer subagent unavailable)
If @matcha-reviewer cannot run (rate limit, timeout, or provider error), do not fail the gate. Run the review inline using the tier checklist above (L0/L1/L2/L3) and render the same verdict format (Risk Tier, Scope, Verdict, counts, file:line evidence). Then validate with matcha_review_validate. A self-review is weaker than an independent gate, so note Reviewer: inline (fallback) in the verdict.
Risk-Based Routing
Not all code needs the same review. Matcha auto-detects risk tier and routes accordingly:
| Tier | Risk | What | Review |
|---|---|---|---|
| L0 | Disposable | Spikes, scripts, temp | Output check only |
| L1 | Low | Copy, fixtures, UI text | Lint + typecheck |
| L2 | Product Logic | Features, API, business logic | Full review (9 categories) |
| L3 | High Risk | Auth, payments, DB, crypto | Expert review + threat model |
Auto-Detection
Risk tier is detected via trigger packs — domain-specific signal rules.
Default signals (no pack loaded):
| Signal | Tier |
|---|---|
| Files in security-sensitive paths | L3 (use trigger pack for specifics) |
| Keywords: credentials, tokens, secrets | L3 |
| Database schema changes | L3 |
| New public API endpoints | L2 |
| Test files, fixtures, mocks | L1 |
| Documentation, comments | L1 |
| Files in disposable paths | L0 |
// matcha:explain with justification |
L0 |
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.
- 9d ago Changed · +4 lines 17a50f5321ce
- 24d ago Changed 26d5db523243
- 26d ago Changed · +7 lines c8b8a720c07e
- 2mo ago First seen · 151 lines · 20 tokens per session scan A ad7830582826
matcha:review is a command published in the GitHub repository plumpslabs/matcha (2 stars, last pushed 4d ago), licensed MIT. It adds 20 tokens to every session and 1,734 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
pr-review
Review a PR across multiple dimensions (description, code, scan, tests).
handle-pr-comments
Fetch all PR comments, classify by severity, and present for user approval before acting.
self-review
Review only changed files in this branch Your primary goal is to provide valuable, trustworthy feedback while avoiding false positives and low-impact commentary.
refactor-clean
Refactor provided code for cleanliness, maintainability, and alignment with SOLID principles and modern best practices — no over-engineering.
audit-agents-skills
Audit quality of agents, skills, and commands in a Claude Code project.
review
Review the current diff against project memory.