codeact

A coding workflow that uses Python tools to perform tasks when the user explicitly asks for it.

In plain words
What is it for?
Running Python code, inspecting available project tools, and carrying out tasks through the permitted tool interfaces.
Why use it?
It gives the agent a defined way to choose and run approved Python tools instead of handling those tasks informally.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/gradion-ai/ipybox/codeact
Any agent
npx skills add gradion-ai/ipybox --skill codeact
Clone the repo
git clone --depth 1 https://github.com/gradion-ai/ipybox

Made for: Claude Code, Codex.

Per session 30 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 495 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00030 $0.00495
Opus 5 $0.00015 $0.00247
Sonnet 5 $0.00006 $0.00099
Haiku 4.5 $0.00003 $0.00049

Measured 2d ago against content hash 2cf1bf4a7c39, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codeact 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.

plugin/skills/codeact/SKILL.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Use Python tools to perform tasks.

You must use the execute_ipython_cell tool of the ipybox MCP server for executing Python code.

All operations must follow the tool usage restrictions and workflows defined below.

Tool Directories

The gentools/ and mcptools/ directories are in the working directory shown in your <env> block. All paths are relative to the working directory, NOT to this skill's base directory.

Tool Usage Restrictions

You are restricted to these tools only:

Python Tools

  • Functions in mcptools/<category>/<tool>.py (use run_parsed if defined, otherwise run)
  • Functions in gentools/<category>/<tool>/api.py

ipybox MCP Server Tools

  • execute_ipython_cell - Execute Python code
  • reset - Reset the IPython kernel

Claude Code Filesystem Tools

  • All filesystem tools for reading, writing files, and listing directories.

Workflow

1. Python Tool Selection

  1. List available categories in gentools/ and mcptools/
  2. List available tools in relevant categories
  3. Read tool files to understand interfaces and parameters.

2. Python Tool Priority

  1. Search gentools package first
  2. If not found, search mcptools package
  3. If no appropriate tool exists, generate custom code

3. Code Generation and Python Tool Chaining

  • Generate code that uses selected Python tools as argument for execute_ipython_cell.
  • Chain Python tools in the generated code if the structured output of one tool can be used as input for another tool.

4. Code Execution

  • Use the execute_ipython_cell for Python code execution
  • Print only required information, not intermediate results
  • Store intermediate results in variables

Output Parsers

When generating output parsers for Python tools in the mcptools package, see references/output-parsers.md.

Saving Code Actions

To save executed code as a reusable gentools tool, see references/saving-codeacts.md.

Read the full file on GitHub · 66 lines

Files

What ships with it

2 files 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.

Changes

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.

  1. 2d ago First seen · 66 lines · 30 tokens per session scan A 2cf1bf4a7c39

Subscribe to this mod's changes

codeact is a skill published in the GitHub repository gradion-ai/ipybox (73 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 30 tokens to every session and 495 once invoked, about $0.0002 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.

Related

Other skills, from other repositories