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 skills add Adelie-Squad/solosquad --skill primitive-reviewgit clone --depth 1 https://github.com/Adelie-Squad/solosquadWrote 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/skills/adelie-squad/solosquad/primitive-review)<a href="https://agentmods.dev/skills/adelie-squad/solosquad/primitive-review"><img src="https://agentmods.dev/badge/skills/adelie-squad/solosquad/primitive-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/skills/adelie-squad/solosquad/primitive-review"><img src="https://agentmods.dev/badge/skills/adelie-squad/solosquad/primitive-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 | $0.00076 | $0.00784 |
| Opus 5 | $0.00038 | $0.00392 |
| Sonnet 5 | $0.00015 | $0.00157 |
| Haiku 4.5 | $0.00008 | $0.00078 |
Grade A, and why
primitive-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.
What it actually says
Asset Review Skill
정적 validate(kebab id·순환·참조 무결성)가 잡지 못하는 품질·판단 문제를 LLM 관점으로 검토한다.
CLI solosquad <manager> review 를 대체하는 대화형 표면 — review 는 비결정적 출력이라 CLI/CI 게이트가
아니라 Chief 와의 대화에서 수행한다(v1.3.2 conversational-first 결정).
입력
- kind: skill | agent | workflow | goal | cron
- id/경로: 검토 대상 (예:
product/product-manager,workflows/market-research/workflow.yaml) - (선택) 정적 validate 결과 — 이미 알려진 finding 은 반복하지 않는다.
절차
- 대상 정의를 읽는다(
solosquad agent show <id>/workflow show/ 파일 직접). - 먼저
solosquad <manager> validate로 정적 검증을 돌려 그 결과를 컨텍스트로 둔다. - kind 별 렌즈로 판단 문제만 짚는다(정적 규칙이 잡는 건 제외):
- skill — description 이 3인칭·구체적이라 라우팅 가능한가? process 가 실행 가능한가? 단일 목적인가?
- agent — 팀 내 다른 actor 와 역할이 구별되나? 명확한 escalation/handoff 경로가 있나? collaborator 가 과배선 아닌가?
- workflow — 각 stage 의 exit_criteria 가 실제로 측정 가능한가? handoff 순서가 합리적인가? fixed/agentic 선택이 맞나?
- goal — success metric 이 객관적·판정 가능한가? 가드레일(budget/cycles)이 적정한가? modifiable_paths 가 좁게 잡혔나?
- cron — cadence 가 prompt 의도와 맞나? channel 이 kind 에 맞나? prompt 가 무인 실행에 자족적인가?
- 제안을 severity 로 분류해 제시: blocker(고쳐야 함) / improvement(권장) / nit(사소).
출력
- 한 줄 요약 + severity 태그된 actionable 제안 목록.
- anti-sycophancy: 문제가 없으면 "문제 없음"이라고 분명히 말한다 — 억지 제안 금지.
비범위
- 정적 규칙으로 잡히는 것(id 형식·순환·깨진 참조)은
validate의 몫. 여기서 반복하지 않는다. - 자동 수정은 하지 않는다 — 제안만. 적용은 사용자/author 루프가 결정.
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 First seen · 52 lines · 76 tokens per session scan A 3eb6d3486fc7
primitive-review is a skill published in the GitHub repository Adelie-Squad/solosquad (19 stars, last pushed 19d ago), licensed MIT. It adds 76 tokens to every session and 784 once invoked, about $0.0004 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 skills, from other repositories
karpathy-coding-principles
Use when andrej Karpathy's 4 coding principles — think before coding, simplicity first, surgical changes, goal-driven execution. Use when coding, reviewing code quality, reducing overengineering,.
review
Structured code review with parallel audit agents, confidence-scored triage, and optional auto-fix. Examines uncommitted changes, staged diffs, commit ranges, or specific paths. Produces a tiered report (MUST-FIX / RECOMMENDED / NIT) backed by evidence, then optionally applies fixes with verification.
debug
Systematic bug investigation with a five-phase framework: reproduce, narrow, diagnose, fix, verify. Supports automated regression bisect via --regression flag. Produces a structured debug report with root cause analysis, regression test, and CQ/Q self-evaluations.
karpathy-guidelines
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
refactor
Refactor existing code by extracting helpers, splitting files, removing duplication, or untangling dependencies. Supports resumable and batch runs. Use for structural changes that preserve behavior; use zuvo:build for new features.
integrity-check
End-to-end 8-dimension verification — requirements through delivery, all aligned and correct.