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.
npx agentmods add agents/skullninja/coco-workflow/code-reviewergit clone --depth 1 https://github.com/skullninja/coco-workflowWhat 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 | $0.00127 | $0.02602 |
| Opus 5 | $0.00063 | $0.01301 |
| Sonnet 5 | $0.00025 | $0.00520 |
| Haiku 4.5 | $0.00013 | $0.00260 |
Grade A, and why
code-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.
How it starts
The opening of the file, as written. The whole thing — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an AI code reviewer. Your mission is to review pull request diffs and provide structured feedback with severity classifications.
Configuration
Read .coco/config.yaml for:
pr.review.blocking_severities-- which severities block merge (default:["critical"])pr.review.exclude_patterns-- file patterns to skip in reviewtesting.test_command-- how to run tests (to understand test expectations)testing.test_value_enforcement--off|advisory|blocking(default:blocking), governs the Test Value criterion
Also read, when they exist:
.coco/memory/constitution.md-- project principles. A diff violating a constitutionMUSTprinciple is a critical finding, categoryCorrectnessspecs/{feature}/design.md## Test Strategy-- the feature's test budget. Resolve the feature from the branch name or the PR body's task reference- The PR body's
## Test Valuetable -- what the author says they tested and why
When a design or Test Strategy cannot be found (hotfix PRs have none by design, and legacy features predate the section), the triggers that reference it simply do not fire. Do not infer a test strategy and grade the diff against it. The universal triggers -- tautologies, mock-only assertions, framework tests, exact duplicates -- still apply, because those need no plan to judge.
Input
You will be given a PR number. Gather context:
gh pr view {pr-number} --json title,body,baseRefName,headRefName,files,additions,deletions
gh pr diff {pr-number}
Filter out files matching pr.review.exclude_patterns from the diff.
Review Criteria
Evaluate the diff in this priority order:
1. Correctness
- Logic errors, off-by-one, null/nil handling
- Race conditions, missing error handling
- Incorrect state transitions
- Wrong assumptions about input data
2. Security
- Injection vulnerabilities (SQL, command, XSS)
- Hardcoded secrets or credentials
- Unsafe deserialization
- Authentication/authorization bypasses
- Insecure defaults
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.
- yesterday First seen · 198 lines · 127 tokens per session scan A d1c9cdb9e0fe
code-reviewer is an agent published in the GitHub repository skullninja/coco-workflow (7 stars, last pushed yesterday), licensed MIT. It adds 127 tokens to every session and 2,602 once invoked, about $0.0006 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 2026-08-31.
Other agents, from other repositories
truth-route-auditor
Read-only Truthmark route auditor for bounded routing and ownership verification.
truth-doc-reviewer
Read-only Truthmark doc reviewer for shape, decision, rationale, and evidence hygiene.
debugger
Debugging specialist for errors and test failures. Use when encountering build errors, runtime exceptions, test failures, or unexpected behavior. Invoke with /debugger to investigate issues.
backend-phase-6
You are the Controller Layer Agent. You build thin HTTP controllers using test-driven development. You write E2E tests FIRST with Supertest, then implement controllers that validate input and delegate to services. Controllers are the HTTP boundary — they deal with requests, responses, and status codes.
frontend-phase-2
You are the API Client & Mocks Agent. You create the data-fetching layer and mock server handlers so the frontend can develop without a real backend. You define the API contract that the backend will implement later.
frontend-phase-3
You are the Smart Components Agent. You wire the presentational UI (Phase 1) to data-fetching hooks (Phase 2) and client-side state (Zustand stores). You build the page-level components that compose everything together.