Search

Full-text search across content, users, and spaces, plus AI-powered Q&A

The search module provides server-side access to Sublay’s search and AI capabilities. Use it to build search APIs, power recommendation engines, or answer natural-language questions about your content.

---

searchContent

Searches entities by full-text query, optionally scoped to a space.

const { data, pagination } = await sublay.search.searchContent({
  query: "getting started guide",
  spaceId: "spc_abc123",
  sourceTypes: ["entity", "comment"],
  limit: 10,
});

query · string · required

The search query string.

sourceTypes · string[]

Which content types to search: any of "entity", "comment", "message".

spaceId · string

Scope the search to a specific space.

conversationId · string

Scope the search to a specific conversation.

limit · number

Maximum number of results to return.

spaceReputationId · string

Opts each result’s author into a spaceReputation number. Accepts a space <uuid>, "none" (the project-general bucket), or "context" (the space derived from the request context). The empty string and the legacy general / null aliases are rejected. See Reputation.

spaceReputationDescendants · boolean

Only honored alongside an explicit space <uuid>. When true, spaceReputation is the subtree sum — the named space plus all of its descendants.

Returns — Promise<PaginatedResponse<Entity>>

---

searchUsers

Searches for users by name or username.

const { data, pagination } = await sublay.search.searchUsers({
  query: "alice",
  limit: 10,
});

query · string · required

The search query string.

limit · number

Maximum number of results to return.

Returns — Promise<PaginatedResponse<User>>

---

searchSpaces

Searches for spaces by name or description.

const { data, pagination } = await sublay.search.searchSpaces({
  query: "photography",
  limit: 10,
});

query · string · required

The search query string.

limit · number

Maximum number of results to return.

Returns — Promise<PaginatedResponse<Space>>

---

askContent

Asks a natural-language question about your content and returns an AI-generated answer with source references.

This feature requires semantic search to be enabled for your project. See the Semantic Search & AI guide for setup instructions.

const { answer, sources } = await sublay.search.askContent({
  query: "What are the best practices for error handling?",
  spaceId: "spc_abc123",
});

// answer: "Based on the content in this space, best practices include..."
// sources: [{ entityId: "ent_xyz", title: "Error Handling Guide" }, ...]

query · string · required

The natural-language question to answer.

sourceTypes · string[]

Which content types to draw from: any of "entity", "comment", "message".

spaceId · string

Scope the search to content within a specific space.

conversationId · string

Scope the search to a specific conversation.

limit · number

Maximum number of sources to draw from.

spaceReputationId · string

Opts each source’s author into a spaceReputation number. Accepts a space <uuid>, "none" (the project-general bucket), or "context" (the space derived from the request context). The empty string and the legacy general / null aliases are rejected. See Reputation.

spaceReputationDescendants · boolean

Only honored alongside an explicit space <uuid>. When true, spaceReputation is the subtree sum — the named space plus all of its descendants.

Returns — Promise<{ answer: string; sources?: Array<{ entityId: string; title: string | null }> }>