<< All versions
Skill v23.0.1
currentAutomated scan100/100ax-llm/ax/ax-python-playbook
+3 new
──Details
PublishedJuly 20, 2026 at 03:44 AM
Content Hashsha256:4f8e4c13cf31162f...
Git SHA8c45fb74ce22
Bump Typepatch
──Files
Files (1 file, 2.1 KB)
SKILL.md2.1 KBactive
SKILL.md · 49 lines · 2.1 KB
name: "ax-python-playbook" description: "Use when writing Python code with axllm for the playbook() context-engineering surface, agent-bound verified evolution, run-end learning, online updates, and rendering a playbook into a program." version: "23.0.1"
Ax Playbook 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
- Grow an evolving context playbook for a program or agent stage with playbook().
- Attach a seed playbook to an agent and learn bounded avoidance rules from run-end failure signals.
- Use the agent-bound playbook evolve method to mine grounded weaknesses with verification and exact rollback.
- Refine a playbook online from live feedback or offline from labeled examples.
- Render or persist a playbook and inject it into a program context.
Package Facts
- Language: Python.
- Package:
axllm. - Package API docs:
API.mdandaxir-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 ax, playbookprogram = ax("question:string -> answer:string")pb = playbook(program, {"studentAI": llm})pb.evolve(examples, metric_fn)
Relevant API Surface
- 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-apiexamples only when the user explicitly has provider credentials available. - Use
no-keyexamples 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.