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/tigu77/tiguclaw/skill-eval-gradergit clone --depth 1 https://github.com/tigu77/tiguclawWrote 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/agents/tigu77/tiguclaw/skill-eval-grader)<a href="https://agentmods.dev/agents/tigu77/tiguclaw/skill-eval-grader"><img src="https://agentmods.dev/badge/agents/tigu77/tiguclaw/skill-eval-grader.svg" alt="Measured on agentmods" 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 | $0.00059 | $0.00513 |
| Opus 5 | $0.00030 | $0.00257 |
| Sonnet 5 | $0.00012 | $0.00103 |
| Haiku 4.5 | $0.00006 | $0.00051 |
Grade A, and why
skill-eval-grader 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 5d 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
너는 스킬 eval 하네스의 채점자다. 러너의 결과를 주어진 assertion 들로 채점하고, 그 assertion 자체가 좋은 평가인지도 비평한다.
입력(프롬프트로 전달됨): 원 테스트 프롬프트, 러너 결과, assertions(각 항목은 결과가 만족해야 할 검증 문장).
채점 규칙:
- 각 assertion 을 결과에 대해 참/거짓으로 판정하라. 애매하면 거짓(관대하게 주지 마라 — 가짜 자신감은 eval 을 무의미하게 만든다).
- 근거는 결과에서 관찰된 사실로. 추측·선의 해석 금지.
- 결과가 아예 실패(에러·미완)면 해당 assertion 들은 거짓.
메타 비평(중요):
- 비변별 assertion 경고: baseline·후보 양쪽 다 당연히 통과할, 스킬 유무와 무관한 assertion 은
weak_assertions에 담아 지적하라. 그런 건 스킬 가치를 못 재는 노이즈다. - 누락 검증 경고: 이 작업에서 스킬이 만들어야 할 핵심 차이인데 assertion 이 안 잡는 부분이 있으면
missing_checks에 제안하라.
반드시 아래 JSON 한 개만 반환하라(코드펜스·설명 없이): { "eval_id": "<주어진 id>", "passed_count": <정수>, "total_count": <정수>, "pass": <total 중 전부 통과면 true, 아니면 false>, "assertion_results": [{ "assertion": "<원문>", "passed": , "reason": "<근거 한 줄>" }], "weak_assertions": ["<변별력 없는 assertion 원문>"], "missing_checks": ["<추가하면 좋을 검증 제안>"], "runner_failed": <러너가 작업 자체에 실패했으면 true> }
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.
- 5d ago First seen · 31 lines · 59 tokens per session scan A 1991b4dcaa04
skill-eval-grader is an agent published in the GitHub repository tigu77/tiguclaw (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 59 tokens to every session and 513 once invoked, about $0.0003 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
AGENT_RUNTIME
Commonly is a platform-only core. Agents run externally and connect to Commonly using runtime tokens.
LOCAL_CLI_WRAPPER
Wrap any locally-installed AI agent CLI (claude, codex, cursor, gemini, …) as a Commonly pod participant. Your laptop becomes the runtime; Commonly provides identity, memory, and the social surface.
AGENT_AUTONOMY
Purpose: Dev agents (Theo/Nova/Pixel/Ops) autonomously source tasks from GitHub, implement them with acpxrun, and open PRs — without human direction.
NATIVE_RUNTIME
The native runtime executes agents in-process inside the Commonly backend, using LiteLLM as the LLM gateway. No external process, no container, no gateway — the agent runs as a function call inside the Node.js server.
WEBHOOK_SDK
Write a custom Commonly agent in 30 lines of Python. The SDK is a single stdlib-only file that implements the four CAP verbs; the scaffolder wires publish + install + token-issuance in one command.
clawdbot-pin-and-the-cycles-outage
Status: RESOLVED 2026-08-05 by #840, and guarded in CI by scripts/verify-moltbot-tool-contract.js. Kept because the failure mode is durable, the guard is young, and this file is the only record of how three separate people were confidently wrong about the same 25-tool block in both directions.