There are several possibilities for issues encountered:

Network Blocking

Fetching RSS data and calling AI interfaces both require sending network requests. If a request fails or returns a timeout, it might be blocked. In such cases, using a proxy can be a solution:

Excessive Execution Time

You can adjust the following parameters:

Max Items Per RSS Feed indicates how many articles are fetched from each reading source at a time. The larger this value, the more articles to summarize, and the slower the speed. Therefore, you can try reducing this value.

Max Summarize API Concurrency represents how many summary APIs are called simultaneously. The smaller this value, the slower the speed. Therefore, you can try increasing this value. However, the upper limit depends on the limitations of the AI Provider, such as Moonshot AI, which currently only allows a concurrency of 3.

Additionally, AI Providers generally limit the maximum number of calls within a minute, so you might also need to adjust the API Retry related parameters to avoid ultimate article summary failures:

If you feel the default values are too high causing lengthy retries, you can try reducing these two values.

Unknown Raycast Error

If you are manually generating a digest and temporarily close the Raycast window, it may occasionally interrupt without throwing any errors. This issue needs to be fixed by Raycast officially. If you encounter this, one solution is to keep the Raycast window open until the generation is complete. The best solution is to use the auto-generation feature, as detailed in this document.

Other

If there are other issues, invoke Actions in any command and click Bug Report to provide feedback, or directly DM me on 𝕏(Twitter) (opens in a new tab).