nlqdb

Analytical memory for AI agents.

Give your agent a real Postgres database as memory — typed rows it writes as it learns, queried in English.

Connect it to your agent.

Install in your agent.

  • Claude: Settings → Connectors → Add custom connector — paste the URL.
  • Windsurf: Cascade → MCPs → Configure, or ~/.codeium/windsurf/mcp_config.json.
  • Zed: Agent Panel → Add Custom Server, or ~/.config/zed/settings.json.

Anonymous — the configs ship with the pk_live_REPLACE_ME placeholder.

Or spin one up yourself —

Spin up a database from a sentence.

Anonymous — no sign-in. Your DB lives 72h; sign in (always free) to keep it.

What your agent does with it.

Your agent writes a row, then asks a question.

1 · Remember

nlqdb_remember({
  category: "user_preference",
  content: "prefers concise replies",
  user_id: "u_412"
})

→ { id: 4129, ok: true }

The row lands in agent_memory — typed columns, not an opaque vector blob.

2 · Recall

nlqdb_recall({
  question: "feedback per user, ranked by count this month"
})

Compiled SQL (the exact statement nlqdb ran):

SELECT user_id, COUNT(*) AS feedback
  FROM agent_memory
 WHERE category = 'user_preference'
   AND created_at >= date_trunc('month', now())
 GROUP BY user_id
 ORDER BY feedback DESC;
User ID Feedback
u_412 14
u_207 9
u_388 6
u_104 4

Five things — or one tool call.

  • Vector store top-k
  • Per-agent SQLite
  • Embedding cache
  • Schema design
  • Re-rank pipeline
nlqdb_recall(…)
Connect it to your agent →

Not building an agent? nlqdb is also a natural-language database you embed in any app — describe data, drop one tag →