Logeion Agentic Executor — Development skill for Claude Code

---

AI agents are powerful thinkers but poor actors. They call one tool, wait for a result, call another tool, wait again. Every step is a round-trip.

**Agentic Executor** gives the agent a Python interpreter and a set of high-performance commands. Instead of ten tool calls, it writes one script. Loops, conditionals, error handling — all in a single thought.

# One thought. Ten files. Zero round-trips.
python -m agentic_executor.cli --thought-stdin <<'EOF'
files = execute('search', {'pattern': '*.py'})
for f in files['data']['files']:
    result = execute('read', {'path': f, 'lines': 5})
    if result['success']:
        print(f"=== {f} ===")
        print('\n'.join(result['data']['lines']))
EOF

---

Install

pip install agentic-executor

No dependencies. Works immediately with Python's built-in `ast` for Python files. For full multi-language support (20+ languages via tree-sitter):

pip install "agentic-executor[languages]"

---

Activate the /exec skill

After installing, deploy the `/exec` skill to your AI agent of choice:

exec-install                       # Claude Code  (default)
exec-install --agent cursor        # Cursor
exec-install --agent windsurf      # Windsurf
exec-install --agent kiro          # Kiro
exec-install --agent codex         # Codex
exec-install --agent copilot       # GitHub Copilot
exec-install --agent opencode      # OpenCode
exec-install --agent aider         # Aider
exec-install