Search Content
Searches news, posts, repositories, sites, and papers published by or about one company (ownerEntityId) or person (ownerPersonId) from a plain-English query; omit both owners to search every owner’s content. Any signed-in user may call this.
Authentication
Your aVenture API key (https://aventure.vc/settings/api-keys) or OAuth access token
Query parameters
Company whose publications to read; mutually exclusive with ownerPersonId.
Person whose publications to read; mutually exclusive with ownerEntityId.
Zero-based page index (0..N)
Sorting criteria in the format: property,(asc|desc). Default sort order is ascending. Multiple sort criteria are supported.
Request
Plain-English search request.
Optional supported model id for the written answer, which also supplies the page heading, title, and description. Leave the field out for the server’s model; a value that is not a supported model id answers 422.
Cache policy. use reuses earlier search work; bypass and refresh compute fresh results and are honored only for accounts with private-data access.
auto refreshes date-sensitive queries before cache admission; cached permits ordinary cache reuse. Explicit admitted bypass or refresh wins.
Search strategy to run. Defaults to auto: the fast name layer runs first (a single-token query as keyword search, a multiword query as an exact normalized name) and the language-model planner runs only on a miss, so name lookups never pay for planning. Any other value forces exactly that strategy: exact and keyword skip planner model resolution and its rate limit. Entity and person natural-search take every value but hybrid; news and federated search accept only auto and keyword; content search accepts auto, keyword, semantic, and hybrid (semantic and hybrid only for one entity’s or person’s content).
Optional supported model id for planning the search. Leave the field out for the server’s model; a value that is not a supported model id answers 422.
Model catalog revision this search uses for its model defaults and inference policy; null reads the current revision. Callers without private-data access always read the current revision. An unknown revision returns 404.
Optional reasoning effort for every model call this search makes: the query planner and the synthesis answer. Each call sends the nearest level its model supports, the lower one on a tie. Accounts without private-data access are capped at medium; higher values are lowered.
Search catalog revision this search uses for its global search policy and vocabulary; null reads the current revision. Callers without private-data access always read the current revision. An unknown revision returns 404.
Runtime policy revision this search uses for the serving deployment’s execution budgets; null reads that deployment’s current revision. Callers without private-data access always read the current revision. An unknown revision returns 404.
Response headers
Requests allowed per client IP in the current rate-limit window.
Requests left before the tightest applicable rate-limit bucket rejects.
Seconds until the exhausted rate-limit bucket admits another request; 0 when none is exhausted.