Code Execution

This lets Claude run Python locally to handle bulk operations without burning through your token budget. When you're refactoring 50 files or auditing a hundred modules, it executes everything on your machine and returns just the summary. The APIs cover filesystem operations, code analysis (metadata only, not full source), transformations like renaming identifiers across your codebase, and basic git commands. The token savings are legitimate: a 100-file operation drops from 150K tokens to around 1K. It activates automatically when you mention bulk work or efficiency, though you can also import the runtime directly. The tradeoff is setup and trusting local execution, but if you're doing repetitive operations across many files, it's the difference between hitting rate limits and getting work done.

skills.sh installs: 293
GitHub stars: 693
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add mhattingpete/claude-skills-marketplace --skill code-execution --agent claude-code

Installs into .claude/skills of the current project.

Files

Code Execution

Execute Python locally with API access. 90-99% token savings for bulk operations.

When to Use

Bulk operations (10+ files)
Complex multi-step workflows
Iterative processing across many files
User mentions efficiency/performance

How to Use

Use direct Python imports in Claude Code:

from execution_runtime import fs, code, transform, git

# Code analysis (metadata only!)
functions = code.find_functions('app.py', pattern='handle_.*')

# File operations
code_block = fs.copy_lines('source.py', 10, 20)
fs.paste_code('target.py', 50, code_block)

# Bulk transformations
result = transform.rename_identifier('.', 'oldName', 'newName', '**/*.py')

# Git operations
git.git_add(['.'])
git.git_commit('feat: refactor code')

If not installed: Run ~/.claude/plugins/marketplaces/mhattingpete-claude-skills/execution-runtime/setup.sh