> ## Documentation Index > Fetch the complete documentation index at: https://docs.asymptotelabs.ai/llms.txt > Use this file to discover all available pages before exploring further. # ChatGPT > Asymptote support details for ChatGPT browser chat telemetry ## Runtime overview Asymptote collects ChatGPT chat telemetry through the optional [browser extension](/runtimes/browser-extension). Every turn on chatgpt.com becomes prompt and response events in the local runtime log, attributed to the `chatgpt_web` harness so it stays separate from Codex CLI and other agents. Both `chatgpt.com` and the older `chat.openai.com` host are covered. Support is in beta. The ChatGPT chat API is private and undocumented, so a change on the site can interrupt capture until the adapter is updated. ## Prerequisites * The Beacon endpoint is installed and running, and the browser extension is loaded. See [Browser Extension](/runtimes/browser-extension) for both steps. * chatgpt.com is enabled in the extension's options page. It is enabled by default. ## Collection path The extension's main-world interceptor tees the streamed response from the message-send endpoint, `POST /backend-api/f/conversation`. The ChatGPT adapter reads the site's `delta_encoding: v1` Server-Sent Event stream, where each frame patches a running document rather than carrying a finished message, and rebuilds one turn from it: | Stream signal | What Asymptote takes from it | | --------------------------------------------------------- | ------------------------------------------------------------------------- | | The outbound request body | Prompt text, the input messages, and the requested model | | `conversation_id` in the request body or the resume token | `session.id` and `gen_ai.conversation.id` | | `add`, `append`, `replace`, and `patch` delta frames | Assistant response text, assembled as it streams | | The assistant message metadata | The model that actually answered, which can differ from the requested one | | `message_stream_complete` or `[DONE]` | Turn completion | Only the visible assistant message is tracked. Tool and aside messages addressed to another recipient are skipped, so the recorded response is what the user saw. The service worker then posts the turn to the local collector as OTLP GenAI logs. Events carry `harness.collection_method` of `otlp`. A turn whose stream is cut off before completion is still delivered, recorded as `agent.response` rather than `agent.response.completed`. ## Events One ChatGPT turn produces: | Event | When | | -------------------------- | --------------------------------------------------------------------- | | `prompt.submitted` | The request body carried prompt text or input messages | | `agent.response.completed` | The stream finished normally | | `agent.response` | The stream ended early, so the response is partial | | `tool.invoked` | The turn included web search results, recorded as a `web_search` call | Turns from one conversation share a session id, so a conversation reads as a session in the dashboard even though the site sends no session start or stop signal. ## Telemetry coverage | Area | Support | | ---------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- | | Prompt telemetry | Supported, with full prompt text under `full` and `redacted` retention | | Assistant response telemetry | Supported, including the response text, the requested model, and the model that answered | | Tool telemetry | Partial. Web search activity is captured from the assistant message metadata; other built-in tools are not exposed in a form the adapter can read | | Token usage | Not available. ChatGPT web does not report token counts in its stream | | Cost | Not available. ChatGPT web does not report per-turn cost to the page | | Command and file telemetry | Not applicable. Browser chat performs no local commands or file edits | | Approval telemetry | Not available. ChatGPT does not expose approval decisions to a page script | | Session lifecycle | Partial. Turns are correlated by conversation id; there is no session start or stop signal | | Local JSONL and dashboard | Supported after events reach the collector | | MDM deployment | Not covered by Beacon's MDM assets. Distribute through browser policy instead | Token usage is the main difference from [Claude.ai](/runtimes/claude-web), which does report counts. Reports that roll up token usage will show ChatGPT web turns with no token attribution. ## Confirm it is working Send one message on chatgpt.com, then look for the event: ```bash theme={null} grep -c '"chatgpt_web"' ~/.beacon/endpoint/logs/runtime.jsonl ``` To read the most recent one: ```bash theme={null} grep '"chatgpt_web"' ~/.beacon/endpoint/logs/runtime.jsonl | tail -1 | jq '{action: .event.action, harness: .harness.name, model, session: .session.id}' ``` On a system-mode install the log lives at `/var/log/beacon-agent/runtime.jsonl` instead, and reading it needs `sudo`. If nothing arrives, work through the [troubleshooting steps](/runtimes/browser-extension#troubleshooting) on the extension page. ## Limits * **The stream format is private.** The adapter is written defensively and degrades to a partial turn rather than failing, but a site change can still stop capture until the adapter is updated. * **Handed-off streams are not captured yet.** ChatGPT sometimes moves a response to a separate resume endpoint instead of streaming it inline. Those turns are missed today; capturing them is a planned follow-up. * **No token usage.** Nothing in the stream carries token counts, and Asymptote does not estimate them. * **Only the chat stream is read.** Uploaded files, canvas documents, and content rendered outside the stream are not captured. * **Retention defaults to `full`.** Prompt and response text is retained unless you change the mode in the extension's popup. See [retained content](/runtimes/browser-extension#retained-content). ## Related Install, configure, and verify the extension. The other supported browser chat surface. Return to the runtime support overview. How retained content is handled once it reaches the endpoint.