Editorial illustration for OpenAI releases Euphony, a tool to visualize Harmony chats and Codex logs
OpenAI Launches Euphony: Chat Log Visualization Tool
OpenAI releases Euphony, a tool to visualize Harmony chats and Codex logs
OpenAI just added a new piece to its open‑source toolbox, aimed at anyone who has to sift through raw chat logs or code‑generation sessions. The company’s latest release targets two of its own products—Harmony, the conversational model, and Codex, the code‑focused engine—by turning the often‑messy JSON or JSONL files they emit into something you can actually browse in a browser. Until now, developers have had to write bespoke parsers just to line up timestamps, speakers and code snippets in a readable format.
Euphony sidesteps that step entirely, offering a visual timeline that lays out each exchange without the need for hand‑crafted scripts. Even more, the tool can automatically recognize different log structures, handling four distinct detection modes out of the box. That means less time wrestling with data and more time interpreting what the models actually said or generated.
Below are the key takeaways that sum up what the utility brings to the table.
Key Takeaways - OpenAI open-sourced Euphony, a browser-based visualization tool that converts raw Harmony JSON/JSONL conversations and Codex session JSONL files into structured, browseable conversation timelines -- no custom log parsers needed. - Euphony supports four auto-detection modes: it recognizes lists of Harmony conversations, Codex session files, conversations nested under top-level fields, and falls back to rendering arbitrary data as raw JSON objects. - The tool ships with a rich inspection feature set -- including JMESPath filtering, focus mode (filter by role, recipient, or content type), conversation-level and message-level metadata inspection, grid view for dataset skimming, and an in-browser JSONL editor mode.
- Euphony runs in two modes: a frontend-only mode recommended for static or external hosting, and an optional local backend-assisted mode powered by a FastAPI server that adds remote JSON/JSONL loading, backend translation, and Harmony rendering -- with OpenAI explicitly warning against exposing the backend externally due to SSRF risk. - Euphony is designed to be embeddable: it ships as reusable Web Components (
) compatible with React, Svelte, Vue, and plain HTML, with fully customizable styling via CSS custom properties, and is released under the Apache 2.0 license.
Debugging an AI agent that runs for dozens of steps isn’t like stepping through a regular function. No single stack trace. Instead, developers stare at hundreds of lines of raw JSON, trying to piece together what the model thought at each turn.
OpenAI’s release of Euphony tries to change that. The tool runs in a browser, turning Harmony chat logs and Codex session files into a structured, browseable timeline without requiring custom parsers. It auto‑detects four different log formats, though the exact modes aren’t listed in the summary.
By visualizing the sequence of file reads, API calls, code writes, and revisions, Euphony promises a clearer picture of an agent’s execution path. Whether the interface scales to very large sessions remains unclear; the description mentions “dozens of steps,” but performance on longer runs isn’t specified. The open‑source nature invites community testing, yet practical adoption will depend on how easily developers can integrate it into existing workflows.
For now, Euphony offers a concrete step toward making multi‑step AI debugging more approachable.
Further Reading
Related Reading
Browse all OpenAI news →
Open Source UK PM vows action on Grok's deepfake scandal, Starmer condemns X
Open Source GPT-5 helps mathematicians offload tedious tasks, says Timothy Gowers
Open Source India proposes licensing and royalty rules for AI training by Google, OpenAI
Related Topic OpenAI, a Series F San Francisco startup founded in 2015 by eight pioneers
Business & Startups
Related Topic OpenAI's Codex powers Lovable AI, letting millions create apps from text
LLMs & Generative AI
Same Week 91% of businesses now use video marketing — AI cut the cost of keeping up by 91% too
Market Trends
From Archives Hugging Face releases ml‑intern, an agent that auto‑diagnoses LLM failures
From Archives Scammer Uses AI-Generated MAGA Girl to Grift Men, Cites Pro-Nazi Content Rise
From Archives OpenAI upgrades ChatGPT image model, improves English text rendering
LLMs & Generative AI
From Archives Starbucks ChatGPT app forces users to pick from suggested iced‑coffee options
LLMs & Generative AI
Common Questions Answered
What problem does OpenAI's Euphony tool solve for developers working with Harmony and Codex logs?
Euphony addresses the challenge of parsing complex JSON and JSONL files from AI conversations and code generation sessions by providing a browser-based visualization tool. Instead of manually writing custom log parsers, developers can now easily browse structured conversation timelines and understand the context of AI interactions.
How many auto-detection modes does Euphony support for processing different log formats?
Euphony supports four auto-detection modes that can recognize various log structures, including lists of Harmony conversations, Codex session files, conversations nested under top-level fields, and a fallback mode for rendering arbitrary data as raw JSON objects. This flexibility allows the tool to handle diverse log formats without requiring manual configuration.
Why is visualizing AI agent interactions more challenging compared to traditional debugging?
Unlike traditional software debugging with clear stack traces, AI agent interactions involve complex, multi-step processes that generate hundreds of lines of raw JSON. Developers often struggle to piece together the model's reasoning and decision-making at each step, which is why tools like Euphony are crucial for understanding AI conversation flows.