skills.rest
1154761334
1154761334COMMUNITY
Explore all skills by this author →
GitHub

heterogeneous-agent

Analyze trace data to debug heterogeneous AI agent pipelines.

Updated Apr 3, 2026
Search Tags:#debugging#integration#agent#multi-agent#pipeline#jsonl#trace analysis
One-click install
npx skills add https://github.com/1154761334/it-bidding-copilot --skill heterogeneous-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: heterogeneous-agent
Source: https://github.com/1154761334/it-bidding-copilot/tree/main/frontend/.agents/skills/heterogeneous-agent
Command: npx skills add https://github.com/1154761334/it-bidding-copilot --skill heterogeneous-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides guidance for implementing, troubleshooting, and debugging integrations involving heterogeneous AI agents such as Claude Code, Codex, and external CLI tools, ensuring cohesive multi-agent workflows.

Core Features & Use Cases

  • Integration Troubleshooting: Helps identify and fix bugs in adapter event mappings, transport layers, and subagent threading issues.
  • Development Guidance: Offers comprehensive instructions for developing new agent drivers and debugging streaming JSONL traces.
  • Use Case: When deploying a multi-tool AI system integrating Codex and Claude, use this Skill to pinpoint if tools get stuck or produce orphan messages, and streamline the pipeline.

Quick Start

Use the heterogeneous agent skill to diagnose and fix agent pipeline errors or to understand multi-agent message flow.

Frequently Asked Questions about heterogeneous-agent

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

FAQPage Schema
How do I debug multi-agent pipeline errors when integrating heterogeneous AI agents like Claude Code and Codex?▼

Debug multi-agent pipeline errors by analyzing streaming JSONL trace data to identify event mapping mismatches, transport layer failures, and subagent threading issues across heterogeneous AI agents. This process isolates where tools get stuck or produce orphan messages within complex workflows.

What causes orphan messages in multi-agent workflows using external CLI tools?▼

Orphan messages in multi-agent workflows typically stem from adapter event mapping errors and broken session management between heterogeneous agents. Analyzing JSONL trace data helps pinpoint incomplete event loops and transport layer failures in the multi-tool pipeline.

How do I trace JSONL event flows to fix stuck tools in enterprise AI systems?▼

Trace JSONL event flows to fix stuck tools by parsing the streaming execution data to map event sequences, verify subagent threading, and detect coordination failures. This isolates the exact transport or adapter layer causing the pipeline blockage.

Can I use this approach to develop new agent drivers for multi-platform integrations?▼

Yes, you can use this approach to develop new agent drivers because it provides comprehensive instructions for building adapters, debugging transport layers, and ensuring robust session management across heterogeneous agent types in enterprise AI systems.

What is the best way to troubleshoot adapter event mapping in heterogeneous agent integrations?▼

Troubleshoot adapter event mapping by examining JSONL trace outputs to verify that event sequences align correctly across different agent types. This method exposes mismatches in the transport layer and identifies where subagent threading fails.

Why does my multi-tool AI system produce misaligned event sequences between Codex and Claude?▼

Misaligned event sequences between Codex and Claude occur due to incompatible adapter mappings and transport layer desynchronization. Debugging the JSONL trace data reveals the exact event mapping discrepancies causing the heterogeneous pipeline failure.

Related Skills

Explore More Software Engineering Skills→
duwenjiduwenji

heterogeneous-agent

Diagnose LobeHub heterogeneous agent integration failures across CLI trace, adapter mapping, and persistence stages.

Community
Advanced
ericmjlericmjl
38

agents-md-improver

Propose cohesive patches to AGENTS.md from user corrections.

Community
Advanced
esmevaneesmevane
3

oneiros

Manage persistent agent memory and cognitive context with an event-based SQLite system.

Community
Advanced
betalyrabetalyra
30

effect-uai-modify-output-stream

Format a Stream<TurnEvent> for transport via SSE or JSONL.

Official
Intermediate