Search

Semantic search across your workspace's documents using vector similarity (pgvector + Gemini embeddings).

---

Endpoint

GET /api/v1/search?q=<query>

---

Query parameters

Name	Type	Required	Default	Max	Description
q	string	Yes	—	—	Natural language search query
limit	number	No	5	20	Maximum results to return
threshold	number	No	0.5	—	Minimum similarity score (0–1)

---

Example

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://lorehq.co/api/v1/search?q=how+do+we+deploy+to+production&limit=5"

---

Response

{
  "results": [
    {
      "id": "abc123",
      "title": "Production Deployment Runbook",
      "content": "## Steps\n1. Run migrations...",
      "similarity": 0.94
    }
  ]
}

Results are ordered by similarity score (highest first). Only results above the threshold are returned.

---

How it works

1. Your query is embedded using Google Gemini (gemini-embedding-001, 768 dimensions)
2. The embedding is matched against document vectors using the match_documents Postgres RPC function (pgvector cosine similarity)
3. Results are filtered by workspace and similarity threshold
4. If embedding generation fails, the API returns a 500 error