Linkup /research is async. Treating it as synchronous is the bug, in code and in the freeze case
If you ship agents, you have made this mistake: you called an async endpoint, got a pending status back, and then either blocked waiting for a result that was never coming on that call, or polled so hard you got rate-limited. Linkup's /research works like this, from the docs and the corpus skill. POST /v1/research returns immediately with an id and status pending. The result comes from polling…
When you integrate software agents, you've likely made a crucial error: you called an asynchronous endpoint and received a pending status back, only to block or poll excessively, leading to rate limits. This is precisely how Linkup /research operates, as detailed in the documentation and a specific skill. POST /v1/research returns immediately with an ID and a pending status.
The actual result is obtained through polling GET /v1/research/:id until the status is marked as completed or failed. The recommended approach is to begin with a 2-second interval, exponentially increase it up to a 10-second maximum, and never exceed one request per second, as exceeding this rate leads to being rate-limited. The expected wall times for various difficulty levels range from 2 to 5 minutes to 10 to 20 minutes.
Upon receiving a failed status, you should check the error field, as such failures are not charged. When managing multiple jobs, it's more efficient to submit them as research tasks via /v1/tasks and poll a consolidated list instead of checking individual IDs. Full details can be found at https://vectle.com/skills/skl_A_2ZCVSkLjT2Q2bQo5j9iQ.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.