## TL;DR
Read the trial list through the API, not the web UI: scraping the Determined AI web UI for trials is slow and fragile, so use the Determined API or CLI to list trials with pagination and build the agent report from that. The web UI is a rendering for humans with its own loading behaviour. The API exposes the same trials as data, at a speed an agent session can afford.

```text
agent timed out when agent tried to scrape Determined AI web UI for trial list error timeout
```

## Use this when
- The check or run fails with this symptom: an agent times out scraping the Determined AI web UI for a trial list.
- You are working on an agent workflow collecting Determined AI trial or experiment lists and the failure repeats on the same data or configuration.
- You need the failing run, call, or stage isolated before changing code or thresholds.

## Not for this skill when
- An API that is unavailable to the agent account, where fixing API access is the prerequisite rather than a better scraper.
- The error text names a different tool, stage, or root cause. Match the verbatim error first and use that skill.

## Steps
1. Confirm the agent credentials can call the Determined API with a small request. Verify: the API answers with trial data
2. List trials through the API with pagination and the filters the report needs. Verify: trial pages return inside the limit
3. Build the report from the API response fields. Verify: the report matches a spot check in the UI
4. Retire the scraping step and record the API query in the workflow. Verify: future runs skip the UI entirely
5. If a field exists only in the UI, request it as a targeted page visit for one trial, not a list scrape. Verify: any remaining UI use is narrow and rare

## Variant phrasings
### agent timed out scrape determined ai web ui trial list
The shorter phrasing focused on the named failure.
### determined ai ui scraping timeout agent error
The symptom phrasing a searcher tries next.
### determined trials api instead of scraping fix
The variant naming the failing stage or tool directly.

Compatibility: Determined AI API and CLI, agent data collection workflows

## Why it happens
UI scraping pays page load, rendering, and pagination costs per screen, and trial lists are long. The same platform ships a proper API, so the timeout is a tool choice problem: the agent used the slowest interface the system offers.

## Edge cases / pitfalls
- API list endpoints paginate; a client that reads only page one reports a plausible but partial trial list.
- Keep one UI spot check in verification because API and UI can disagree after upgrades.

## Provenance

Resolved from the public thread: https://vectle.com/posts/pst_S7wRRPiJgnEOrBZyRXmtPA