# Ax Python Agent Optimize

> Use when writing Python code with `axllm` for agent optimization, verified agent-playbook evolution, evaluators, judges, optimizer artifacts, BootstrapFewShot, and GEPA.

- Skill: `ax-llm/ax-python-agent-optimize` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ax-llm/ax-python-agent-optimize`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ax-llm/ax-python-agent-optimize/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: ax-llm (https://skillmd.com/u/ax-llm)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/ax-llm/ax-python-agent-optimize

---

# AxAgent Optimize For Python

This skill helps an agent write Python code with the generated Ax package `axllm`. Use the generated package API, examples, and manifests; do not import TypeScript-only APIs unless you are editing the TypeScript package.

## When To Use

- Optimize an AxAgent or reusable program component.
- Mine grounded weaknesses from failed agent tasks and keep only playbook proposals that pass the verification gate.
- Create evaluator callbacks and persist optimizer artifacts.
- Keep optimization runs bounded by explicit budgets and dataset rows.

## Package Facts

- Language: Python.
- Package: `axllm`.
- Package API docs: `API.md` and `axir-api.json`.
- Capability manifest: `axir-capabilities.json`.
- Runnable examples: `examples/`.
- Real network support: yes.
- Scripted no-key transport support: yes.
- Runtime profiles: `javascript-quickjs`, `python-pyodide`.

## Core Pattern

```python
from axllm import AxGEPA

engine = AxGEPA(reflection_client)
result = engine.optimize(request, evaluator)
```

## Relevant API Surface

- Agents And RLM: `agent`, `AxAgent`, `AxAgent.add_child_agent`
- Optimizers: `optimize`, `playbook`, `AxPlaybook`, `AxBootstrapFewShot`, `AxGEPA`, `OptimizerEngine`, `OptimizerEvaluator`

## Guardrails

- Start from package examples for exact native syntax before inventing a new call shape.
- Use `provider-api` examples only when the user explicitly has provider credentials available.
- Use `no-key` examples for deterministic local checks and provider request mapping.
- Treat AxIR as the source of generated package truth: if package docs disagree with source code, update the compiler and regenerate packages.
- Do not copy repo-maintainer skills from `tools/*/skills/` into user packages.