skills.rest
ax-llm
ax-llmOFFICIAL
Explore all skills by this author →
GitHub

ax-cpp-agent

Write C++ Ax agent code with tools, child delegation, MCP clients, and typed envelopes.

2.9k|186|Updated Feb 23, 2023
Search Tags:#delegation#agents#mcp#c++#runtime state#typed responses#axllm
One-click install
npx skills add https://github.com/ax-llm/ax --skill ax-cpp-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ax-cpp-agent
Source: https://github.com/ax-llm/ax/tree/main/website/static/cpp/.well-known/agent-skills/ax-cpp-agent
Command: npx skills add https://github.com/ax-llm/ax --skill ax-cpp-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you write correct C++ code for Ax agents without guessing the generated package API, runtime envelopes, or delegation patterns.

Core Features & Use Cases

  • Build agents with tools, child agents, and MCP clients.
  • Use clarification, discovery, recall, final, and respond envelopes.
  • Save and restore runtime state for long-running work.
  • Prefer direct response execution when no tools are needed.
  • Example use case: creating a C++ assistant that answers questions, calls local tools, and resumes safely after interruption.

Quick Start

Use this skill to produce a C++ Ax agent example that forwards a question through axllm and returns the typed output.

Frequently Asked Questions about ax-cpp-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a C++ agent with tools and child delegation?▼

Build C++ agents with tools and child delegation by using the axllm package APIs to configure child agents and local tools, matching native syntax to ensure correct runtime execution and typed response envelopes.

How does runtime state persistence work for long-running C++ agents?▼

Runtime state persistence for C++ agents works by saving and restoring runtime state through axllm APIs, allowing long-running work to resume safely after an interruption without losing execution context.

What are typed response envelopes in C++ agent workflows?▼

Typed response envelopes in C++ agent workflows are structured output formats including clarification, discovery, recall, final, and respond types, ensuring agents return predictable, type-safe results from axllm.

Can I use MCP clients with C++ agents in axllm?▼

Yes, you can use MCP clients with C++ agents in axllm to integrate external capabilities, requiring the generated package APIs and capability manifests to match native deployment behavior for correct execution.

When should I use direct-response skipping for no-tool tasks?▼

Use direct-response skipping for no-tool tasks when an agent only needs to answer questions without calling local tools, bypassing unnecessary delegation steps to execute final responses immediately.

Do I need generated axllm package APIs to write C++ agent code?▼

Yes, you need the generated axllm package APIs, examples, capability manifests, and runtime profiles to write C++ agent code, ensuring the output matches native syntax and deployment behavior correctly.

Related Skills

Explore More Software Engineering Skills→
ax-llmax-llm
2.9k

ax-cpp-flow

Build C++ Ax flow programs with axllm native syntax.

Official
Intermediate
ax-llmax-llm
2.9k

ax-go-agent

Build Go agents using the generated Ax package API.

Official
Advanced
ax-llmax-llm
2.9k

ax-python-agent

Build Python agents using the generated axllm package and its typed API.

Official
Intermediate
brendadeeznuts1111brendadeeznuts1111

implementing-jsc-classes-cpp

Implement JavaScript classes in C++ using JavaScriptCore bindings.

Community
Advanced