Monitoring a Codex orchestrator + subagents setup with JSONL traces

Melvin Vivas · X post · 2026-09-06 · Open on X

Topics: AI Agents, Tool Use & MCP, LLMOps, Deployment & Monitoring, AI Dev Tools & Productivity · Level: intermediate

Summary

The creator is testing a Codex setup where GPT-6 Astra orchestrates and Luna models run as subagents. He uses his jsonl-viewer, which supports traces, to watch what the agents are doing. The lesson: in multi-agent setups, check the trace logs to see what each agent is doing.

Key points

• Setup: a strong model (Astra) as orchestrator and a cheaper model (Luna) for subagents in Codex.
• Look at trace logs (JSONL) to follow what the orchestrator and subagents are doing.
• jsonl-viewer's traces support is useful for this kind of monitoring.

Resources mentioned

• donvito/jsonl-viewer · repo · github.com · free
  Open-source tool for viewing and editing .jsonl files (fine-tuning datasets) and agent traces from Codex, Claude Code, Pi and Hermes.
  Also in: jsonl-viewer: Live Traces for Codex Orchestrator and Subagents (Melvin Vivas on X · notes), See live Codex traces with the jsonl-viewer tool (Melvin Vivas on X · notes), Melvin Vivas's open-source AI tools, built with Codex (Melvin Vivas on X · notes), Analyzing Codex Traces with jsonl-viewer (Melvin Vivas on X · notes) and 12 more
• OpenAI Codex · tool · openai.com · paid
  OpenAI's coding agent. In the diagram it writes code, fixes review findings and drives the build loop. The creator also used it to make this video.
  Also in: An agent bot that installs and drives Codex on its own (Melvin Vivas on X · notes), Asking a Coder bot to install Codex (Melvin Vivas on X · notes), Sign in with ChatGPT: Setting Usage Limits for Each App (Melvin Vivas on X · notes), Codex Cloud Environments Must Be Saved & Published Before Use (Melvin Vivas on X · notes) and 240 more

Try this

• Use a JSONL trace viewer to inspect multi-agent runs.