runs

Check a question before running it (Sharpen)

Assesses whether a research question is specific enough to run and returns at most ONE next question, with chips, a You decide default and the assumption taken if skipped. The user’s text is never rewritten: unclear words come back as underline spans. Required questions come first; optional breadth questions (sources to read first, a comparison) follow once ready. An already-specific question is ready with score 1 and no required question. Not charged. Rate-limited per account (30 a minute, 600 a day); 429 with Retry-After. 404 feature_not_enabled where Sharpen is off.

post/v1/assess

Request

  • Base URL: https://api.hyperresearch.ai
  • URL: https://api.hyperresearch.ai/v1/assess
  • Auth: one of:
    • HTTP bearer
    • API key in cookie hr_session

Request body

querystring required
tier'light' | 'deep'
briefSharpenBriefEntry[] (unresolved $ref)

Settled axes (the tags under the box). Skipping a question is an entry with source: "you_decide".

must_readstring[]
metadataobject

The Pulse keys, as on POST /v1/runs: time_window ("last_<N>d", {"last_days": N} or {"start", "end"?}) settles the time axis; pulse (boolean).

Response

Assessment

readyboolean required
scorenumber required
missingstring[] required
settled_by_textstring[] required
titlestring required
assessed_atstring required
pulseboolean

Changes

No recorded changes to this operation across all 1 revision of this API.