/deep-research

Fans out web searches for a question, fetches and cross-checks sources, and produces a cited report.

[Workflow] Runs as a bundled dynamic workflow, distributes work across many subagents, runs in the background.

✅ WHEN TO USE IT?

A thorough, multi-source research task with citations on an open question

/deep-research what are current best practices for LLM prompt caching in production?

Let research run in the background while working on something else

/deep-research compare pricing models of major vector database providers

⛔ WHEN NOT TO?

A quick single lookup of a concrete fact

fanning out across many subagents is unnecessary overhead for one quick lookup

Better: use /web-search for the quick single lookup

SOURCES