Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
Go agentagenticaichatbotdsh-pluginembeddingsevaluationgenerative-aigolangknowledge-basellmmulti-tenantollamaopenaiquestion-answeringragrerankingsemantic-searchvector-searchwiki
import type { SuggestedQuestion } from '@/api/agent/index'
// The knowledge source a suggested question was generated from. Sent with the
// picked question so the agent searches that source before answering; the
// backend treats it as a hint inside the turn's search targets.
export interface QuestionOrigin {
knowledge_base_id: string
knowledge_id?: string
}
// Options bound to one send, carried from the composer to the page that
// issues the request. Keeping them on the send itself (rather than in shared
// "pending" state) means a send that never happens, or a second pick before
// the first one goes out, cannot attach them to the wrong question.
export interface SendMessageOptions {
questionOrigin?: QuestionOrigin
}
// Agent-authored starters have no knowledge source and yield undefined.
export function questionOriginFromSuggestion(
item: Pick<SuggestedQuestion, 'knowledge_base_id' | 'knowledge_id'> | null | undefined,
): QuestionOrigin | undefined {
const knowledgeBaseId = item?.knowledge_base_id?.trim()
if (!item || !knowledgeBaseId) {
return undefined
}
const origin: QuestionOrigin = { knowledge_base_id: knowledgeBaseId }
const knowledgeId = item.knowledge_id?.trim()
if (knowledgeId) {
origin.knowledge_id = knowledgeId
}
return origin
}