# Ax Python LLM

> Use when writing Python code with `axllm` for using the generated Ax package, factory functions, package docs, examples, and API reference.

- Skill: `ax-llm/ax-python-llm` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ax-llm/ax-python-llm`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ax-llm/ax-python-llm/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-llm

---

# Ax LLM Quick Reference 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

- Start a generated-language Ax program from package docs or examples.
- Translate the Ax mental model into the target package without TypeScript-only imports.
- Choose the native package entrypoints for signatures, providers, generators, agents, flows, and optimizers.
- Find ordered or adaptive provider-balancing guidance in the language-specific AI skill.

## 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
import os
from axllm import ai

llm = ai("openai", api_key=os.environ["OPENAI_API_KEY"])
```

## Relevant API Surface

- Signatures: `s`, `f`, `AxSignature`
- AxGen: `ax`, `AxGen`, `run_control`, `AxRunControl`
- AxAI: `ai`, `AxCancellationToken`, `AxAIServiceAbortedError`, `get_supported_ai_models`, `dict[str, str]`, `Callable[[dict[str, str]], dict[str, str]]`, `AIClient.owned_worker_factory`, `AxChatSession`, `closable generator`, `OpenAICompatibleClient`, `OpenAIResponsesClient`, `GoogleGeminiClient`, `AnthropicClient`, `AxUsageContext`, `AxUsageEvent`, `AxUsageObserver`, `set_usage_observer`, `AxRuntimeHooks`, `AxRateLimitInfo`, `AxRateLimiter`, `AxTracer`, `AxMeter`, `AxGlobals`, `set_rate_limiter`, `set_tracer`, `set_meter`, `AxBalancer`, `AxBalancerAdaptiveStrategy`, `AxBalancerStatsStore`, `AxInMemoryBalancerStatsStore`, `create_balancer_route_stats`, `update_balancer_route_stats`, `sample_balancer_route_health`, `MultiServiceRouter`, `ProviderRouter`
- Agents And RLM: `agent`, `AxAgent`, `AxAgent.add_child_agent`
- Flow: `flow`, `AxFlow`, `AxProgram.owned_worker_factory`
- 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.