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/ericwang915/pythonclaw/code_runnernpx skills add ericwang915/PythonClaw --skill code_runnergit clone --depth 1 https://github.com/ericwang915/PythonClawWhat 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 | $0.00058 | $0.00414 |
| Opus 5 | $0.00029 | $0.00207 |
| Sonnet 5 | $0.00012 | $0.00083 |
| Haiku 4.5 | $0.00006 | $0.00041 |
Grade A, and why
code_runner 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.
What it actually says
Code Runner
When to Use
- Run or test Python code snippets
- Evaluate expressions (e.g., "calculate 2^100")
- Prototype logic or verify algorithms
- Execute user-provided Python code safely
When NOT to Use
- Running shell commands or external programs — use
run_command/ terminal - Executing non-Python code (JavaScript, etc.)
- Code requiring heavy I/O, network calls, or persistent state
- Editing or analyzing code without running it
Setup
No API keys or credentials needed. Uses the bundled run_code.py script.
Usage/Commands
python {skill_path}/run_code.py [options]
Input methods:
| Method | Example |
|---|---|
| Inline code | --code "print(2+2)" |
| From file | --file script.py |
| Stdin | Pipe code via stdin |
Options:
| Option | Description | Default |
|---|---|---|
--timeout N |
Max execution time (seconds) | 30 |
--format json |
Output as JSON (stdout, stderr, exit_code) | text |
--no-capture |
Stream output directly | off |
Notes
- Code runs in a subprocess (not eval/exec in the agent process) for safety
- Timeout prevents infinite loops
- Working directory is the project root
- For shell commands or other languages, use the terminal directly
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 59 lines · 58 tokens per session scan A 503a7ad16c59
code_runner is a skill published in the GitHub repository ericwang915/PythonClaw (41 stars, last pushed 24d ago), licensed MIT. It adds 58 tokens to every session and 414 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 2026-08-30.
Other skills, from other repositories
typescript-expert
TypeScript and JavaScript expert with deep knowledge of type-level programming, performance optimization, monorepo management, migration strategies, and modern tooling.
antfu
Anthony Fu's opinionated tooling and conventions for JavaScript/TypeScript projects. Use when setting up new projects, configuring ESLint/Prettier alternatives, monorepos, library publishing, or when the user mentions Anthony Fu's preferences.
conventional-commits
Conventional Commits specification for consistent, machine-readable git commit messages. Use when crafting commit messages, enforcing commitlint rules, or generating changelogs/releases.
tencent-channel-community
腾讯频道(QQ频道)社区管理 skill（CLI 版）。频道创建/设置/搜索/加入/退出，成员管理/禁言/踢人，帖子发布/编辑/删除/移动/搜索，评论/回复/点赞，版块管理，分享链接解析，频道私信，加入设置管理，内容巡检，问答自动回复。涉及腾讯频道、频道帖子、频道成员相关任务时应优先使用。.
migrate-zhin-plugin-runtime
Migrate legacy Zhin.js plugins and projects to the convention-based Plugin Runtime — from usePlugin/getPlugin/addCommand/addMiddleware/addComponent/addTool/addCron/declareConfig/useContext and mutable registries to definePlugin + capability directories + Scope/Token resources. Use this whenever a Zhin plugin fails to…
zhin-audit
Audit Zhin.js monorepo for security vulnerabilities, performance bottlenecks, and architecture issues. Use when asked to "audit code", "check security", "find performance issues", "review architecture", "scan for vulnerabilities", "check memory leaks", or "review code quality".