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 commands/cognitx-leyton/codegraph/critiquegit clone --depth 1 https://github.com/cognitx-leyton/codegraphWhat 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 | $0.00013 | $0.00654 |
| Opus 5 | $0.00006 | $0.00327 |
| Sonnet 5 | $0.00003 | $0.00131 |
| Haiku 4.5 | $0.00001 | $0.00065 |
Grade A, and why
critique 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 2d 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Critique (Step 7)
Plan: $ARGUMENTS
Verify the plan was fully and correctly implemented. This is NOT a code review (that's /review-pr) — this checks completeness against the plan's specification.
Phase 1: Load Plan
Read the plan and extract:
- Tasks — every task listed
- Acceptance Criteria — every checkbox
- Files to Change — every CREATE/UPDATE entry
- Validation commands — test/build commands specified
Phase 2: Task Verification
For each task in the plan:
2.1 File exists / was modified
- CREATE tasks: verify the file exists
- UPDATE tasks: verify the file was modified (check git diff)
2.2 Implementation matches spec
- Read the file and verify the task's requirements are met
- Check that the specified patterns were followed
- Verify the "Implement" bullet points are all addressed
2.3 Codegraph structural verification (where applicable)
If the plan added new classes, functions, endpoints, or edges:
.venv/bin/codegraph query "MATCH (c:Class {name:'<expected>'}) RETURN c.name, c.file"
.venv/bin/codegraph query "MATCH (e:Endpoint) WHERE e.path = '<expected>' RETURN e.method, e.path"
Run /graph-refresh first if needed to ensure the graph reflects current code.
Phase 3: Acceptance Criteria
For each acceptance criterion in the plan:
- Check if it's met by examining the code, tests, or running verification commands
- Mark PASS or FAIL with evidence
Phase 4: Test Verification
cd codegraph
.venv/bin/python -m pytest tests/ -q # all tests pass
python -m compileall codegraph/ -q # byte-compile clean
Verify:
- New tests were written for new code
- Test count increased as expected
- No warnings
Phase 5: Verdict
If ALL criteria pass:
Critique: PASS
--------------
Tasks: {N}/{N} verified
Acceptance criteria: {N}/{N} met
Tests: {N} passed ({M} new)
Ready for: /package
If ANY criteria fail:
Critique: FAIL
--------------
Tasks: {N}/{M} verified — {K} gaps
Acceptance criteria: {N}/{M} met
Gaps:
1. [TASK X] <what's missing>
2. [CRITERION Y] <what's not met>
Action: Run /implement to address the gaps, then /review-pr, /commit, and /critique again.
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.
- 2d ago First seen · 94 lines · 13 tokens per session scan A de1dfc9a2af9
critique is a command published in the GitHub repository cognitx-leyton/codegraph (11 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 13 tokens to every session and 654 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
feedback
Security Design Review — PRD/기획서 기반 보안 의견서·검토 의견서 생성.
compliance
이전 보안 진단 보고서의 Finding들이 패치되었는지 확인하고, 변경된 코드에서 신규 취약점을 탐색합니다.
va
Vulnerability Assessment — 8차원 아키텍처 진단 + Self-Verify + Evidence Verification.
pentest
Penetration Testing — 시나리오 기반 모의해킹 + POC + 라이브 검증 (State Delta 기반).
redteam
Red Team Operations — 인프라 설정 보안 리뷰 + MITRE ATT&CK + Detection Engineering.
verify
Adversarial Verification — 보안 진단 보고서 독립 검증 (Autonomous First).