Skip to main content

ax-cpp-playbook

Use when writing C++ 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.

ソース情報

リポジトリ
ax-llm/ax
ソースの最終更新活動
2026年11月6日 20:24
検出された SKILL.md の言語
英語
スター
2,978
フォーク
201

インストール方法

デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。

ソースファイルを確認

インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。

SKILL.md を表示中

SKILL.md
ソースの指示 · 読み取り専用プレビュー
name
ax-cpp-playbook
description
Use when writing C++ 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
25.4.9
# Ax Playbook For C++ This skill helps an agent write C++ 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: C++. - 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 ```cpp auto program = axllm::ax("question:string -> answer:string"); auto pb = axllm::playbook(program, *llm); pb.evolve(examples, metric_fn); ``` ## Relevant API Surface - Optimizers: `axllm::optimize`, `axllm::playbook`, `axllm::AxPlaybook`, `axllm::AxBootstrapFewShot`, `axllm::AxGEPA`, `axllm::OptimizerEngine`, `axllm::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.
GitHubで見る