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 skills/crisandrews/clawcode/skill-managernpx skills add crisandrews/ClawCode --skill skill-managergit clone --depth 1 https://github.com/crisandrews/ClawCodeWhat 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 | $0.00074 | $0.00838 |
| Opus 5 | $0.00037 | $0.00419 |
| Sonnet 5 | $0.00015 | $0.00168 |
| Haiku 4.5 | $0.00007 | $0.00084 |
Grade A, and why
skill-manager 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Manager
Add community skills to this agent, list what's installed, or remove what you no longer want. A "skill" here is a directory with a SKILL.md — the same format Claude Code natively uses — so the exact same artifact works in either place.
This is a CORE feature. See docs/skill-manager.md for the full reference.
Dispatch
Parse the user's command and route to the correct MCP tool:
| User says | Action |
|---|---|
/agent:skill install <source> |
Call skill_install({ source }) |
/agent:skill install <source> --scope=<plugin|project|user> |
Call skill_install({ source, scope }) |
/agent:skill install <source> --force |
Call skill_install({ source, force: true }) — only after the user confirms an overwrite |
/agent:skill list |
Call skill_list() and print the card |
/agent:skill remove <name> |
Call skill_remove({ name }) (dry run) then ask the user to confirm, then call again with confirm: true |
/agent:skill remove <name> --force |
Call skill_remove({ name, confirm: true }) directly — no confirmation prompt |
Default scope is plugin (./skills/). Users can promote a skill later.
Install flow
- Validate the source (GitHub shorthand, full URL, or local path — the tool parses all three)
- Call
skill_install({ source })— the tool clones, detects format, checks requirements - Print the returned card verbatim
- If
format: openclawwas detected (rejected), DO NOT retry withforce. Suggest/agent:import-skilland stop. - If the skill installed with warnings, remind the user what needs to be set up (missing env var, binary, etc.)
List flow
- Call
skill_list() - Print the card. On messaging channels, compress to one line per skill.
Remove flow
- Call
skill_remove({ name })first — this is a dry run; it reports what WOULD be removed - Ask the user: "Remove from ? [y/N]" — short and clear
- If they confirm, call
skill_remove({ name, confirm: true }) - If they say no, just acknowledge and stop
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 · 69 lines · 74 tokens per session scan A da303f0f7c8d
skill-manager is a skill published in the GitHub repository crisandrews/ClawCode (62 stars, last pushed 2d ago), licensed MIT. It adds 74 tokens to every session and 838 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
proof-checker
Rigorous mathematical proof verification and fixing workflow. Reads a LaTeX proof, identifies gaps via cross-model review (external reviewer backend, ultra reasoning), fixes each gap with full derivations, re-reviews, and generates an audit report. Use when user says "检查证明", "verify proof", "proof check", "审证明"…
citation-audit
Zero-context verification that every bibliographic entry in the paper is real, correctly attributed, and used in a context the cited paper actually supports — catching hallucinated authors, wrong years, fabricated venues, version mismatches, and wrong-context citations. Use when user says "审查引用", "check citations"…
idea-creator
Generate and rank research ideas given a broad direction. Use when user says "找idea", "brainstorm ideas", "generate research ideas", "what can we work on", or wants to explore a research area for publishable directions.
paper-illustration
Generate publication-quality AI illustrations for academic papers using Gemini image generation. Creates architecture diagrams, method illustrations with Claude-supervised iterative refinement loop. Use when user says "生成图表", "画架构图", "AI绘图", "paper illustration", "generate diagram", or needs visual figures for papers.
idea-discovery
Workflow 1: Full idea discovery pipeline to go from a broad research direction to validated, pilot-tested ideas. Use when user says "找idea全流程", "idea discovery pipeline", "从零开始找方向", or wants the complete idea exploration workflow.
paper-illustration-image2
Generate publication-quality academic illustrations through a local Codex app-server bridge that uses Codex native image generation. This is a separate experimental alternative to paper-illustration, intended for Claude Code users who want a GPT-image-style renderer without modifying the original skill.