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

ax-cpp-agent-observability

Diagnose Ax C++ agent issues by interpreting traces, usage accounting, action logs, envelopes, and replay data.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you understand and debug generated Ax C++ agents when you need to inspect traces, usage, action logs, checkpoints, or runtime behavior without guessing what happened.

Core Features & Use Cases

  • Trace inspection: Read agent traces and runtime envelopes to see how a request progressed.
  • Usage accounting: Review token and usage data for cost and performance analysis.
  • Replay and debugging: Investigate loops, tool calls, and failures in production-like runs.
  • Use case: Use this skill when a C++ agent behaves unexpectedly and you need to trace the execution path, inspect actions, and pinpoint the failing step.

Quick Start

Use this skill to inspect a C++ agent trace and explain the runtime envelope, usage metrics, and action log for the failing request.

Frequently Asked Questions about ax-cpp-agent-observability

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

FAQPage Schema
How do I trace execution paths in a generated Ax C++ agent?▼

To trace execution paths in a generated Ax C++ agent, inspect agent traces and runtime envelopes to see how a request progressed. This allows you to review the exact sequence of actions without guessing what happened during the flow.

Why does my C++ agent behave unexpectedly during production runs?▼

A C++ agent behaves unexpectedly due to failing loops, tool calls, or runtime failures. Investigate these issues by replaying production-like runs and inspecting the action log to pinpoint the failing step.

How can I review token usage and costs for Ax C++ agent flows?▼

Review token usage and costs for Ax C++ agent flows by analyzing usage accounting data. This provides detailed metrics for performance analysis and cost evaluation of your generated agent requests.

Do I need specific manifests to debug axllm agent callbacks locally?▼

Debugging axllm agent callbacks locally requires the generated Ax C++ package API, package examples, capability manifests, and runtime profile support for accurate local or networked inspection.

What is the best way to replay and debug failing C++ agent tool calls?▼

The best way to replay and debug failing C++ agent tool calls is to use replay data to investigate loops and failures in production-like runs. This isolates the exact failing step in the action log.

Related Skills

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

ax-java-agent-rlm

Build Java Ax agent runtime code for the dev.axllm:ax package.

Official
Intermediate
ryanmatryanmat

lm-apm

Investigate LogicMonitor APM trace data to diagnose service issues and correlate alerts.

Community
Advanced
ax-llmax-llm
2.9k

ax-python-agent-observability

Diagnose Python Ax agent observability issues in axllm workflows.

Official
Intermediate
ax-llmax-llm
2.9k

ax-rust-agent-observability

Inspect Rust Ax agent traces and runtime envelopes for debugging.

Official
Intermediate