// claude + superagnt_

Can Claude save data between chats?

Yes. Claude's built-in memory keeps facts and preferences across chats, and paid plans can search past chats. For exact records, connect the superagnt_ MCP server: Claude saves rows to a real Postgres database that any later chat reads back word for word.

Memory is for what Claude should know about you; a database is for what you need back exactly, like decisions, lists and leads. In the run below, one chat saved two decisions and a new chat 15 minutes later read them back in 1.1 seconds.

Copy the setup prompt.
Your agent does the rest.

works with any agent+ 8 more

Install superagnt for me. It's the superagnt plugin from the github.com/superagnt/plugins marketplace, plus the superagnt MCP server (on Claude it's a connector I add in my browser; elsewhere I sign in through my browser). The setup guide for each kind of agent is at https://www.superagnt.com/install

or follow the detailed setup instructions

» the base connection already serves your workspace database. Nothing to switch on.

01// on its own, and with superagnt_

What Claude does alone, and what changes once it's connected

job	Claude on its own	with superagnt_
Facts and preferences	Memory saves topics as you chat, and you can say “remember this”. You can read and edit it in settings, and each project has its own.	Keep using Claude's memory for these. superagnt_'s own memory tools are for the agents you host on superagnt_, not for Claude's chats.
Exact records	Memory keeps topics Claude writes as you chat, not rows you can query.	A Postgres table: Claude inserts rows in one chat (`agnt_db_write`) and reads them, word for word, in any later one (`agnt_db_read`).
Searching what you saved	Paid plans can search past chats.	SQL: filter, count, sort and join what you saved, with `agnt_db_sql`.
Across apps	Claude's memory lives in Claude. Claude Code keeps its own (CLAUDE.md files and auto memory, which stays on that machine).	One database answers Claude, Claude Code, ChatGPT and Codex, and the agents you host on superagnt_.

02// a real run

The job, run on superagnt_

Real tool calls on superagnt_'s staging environment, the same calls Claude makes once the server is connected. Each result is what the client received, trimmed where it says so and otherwise unchanged, and each wall time was measured around the call.

the ask

“Make a table for team decisions and save today's two. Then, in a new chat later: what did we decide today?”

01Chat 1: Create a table for decisions656 ms wall timeOct 26, 2026, 12:45 UTC

→ callagnt_db_sql

{
  "action": "apply_migration",
  "name": "create_decisions",
  "query": "create table public.decisions (id bigint generated always as identity primary key, decided_on date not null default current_date, decision text not null, owner text not null)"
}

← result

{
  "success": true
}

02Chat 1: Save the two decisions421 ms wall timeOct 26, 2026, 12:45 UTC

→ callagnt_db_write

{
  "action": "insert",
  "table": "decisions",
  "rows": [
    {
      "decision": "The customer webinar moves to Thursdays at 11am Eastern, starting Nov 5.",
      "owner": "Maya"
    },
    {
      "decision": "Onboarding emails go from five to three; the day-2 email carries the setup video.",
      "owner": "Jon"
    }
  ],
  "return_rows": true
}

← result

{
  "rows": [
    {
      "id": 1,
      "decided_on": "2026-10-26",
      "decision": "The customer webinar moves to Thursdays at 11am Eastern, starting Nov 5.",
      "owner": "Maya"
    },
    {
      "id": 2,
      "decided_on": "2026-10-26",
      "decision": "Onboarding emails go from five to three; the day-2 email carries the setup video.",
      "owner": "Jon"
    }
  ]
}

03Chat 2, 15 minutes later: Read them back1.1 s wall timeOct 26, 2026, 13:01 UTC

→ callagnt_db_read

{
  "action": "select",
  "table": "decisions",
  "order": "id.asc"
}

← result

{
  "rows": [
    {
      "id": 1,
      "decided_on": "2026-10-26",
      "decision": "The customer webinar moves to Thursdays at 11am Eastern, starting Nov 5.",
      "owner": "Maya"
    },
    {
      "id": 2,
      "decided_on": "2026-10-26",
      "decision": "Onboarding emails go from five to three; the day-2 email carries the setup video.",
      "owner": "Jon"
    }
  ]
}

The second chat was a new MCP session with nothing carried over, and the rows came back character for character. The same table is there for ChatGPT, Codex or a hosted agent tomorrow.

03// set it up

Set it up in your client

Claude

01Open the prefilled Add custom connector dialog on claude.ai. One connector covers Claude on the web, Claude Desktop and Cowork.
02Click Add, then approve on the superagnt_ consent screen. It creates your superagnt_ account if you don't have one.
03Ask: “Make a table for our team's decisions and save these: <decisions>”

Free plans can add one custom connector. On Team and Enterprise plans, an owner adds it under Organization settings, Connectors (Anthropic).

Claude Code

01In Claude Code, run /plugin marketplace add superagnt/plugins, then /plugin install superagnt@superagnt for the superagnt_ skills.
02Add superagnt_ to Claude as a connector with the prefilled dialog and click Add. Claude Code signed in with your Claude account picks it up.
03Start a new session.
04Ask: “Make a table for our team's decisions and save these: <decisions>”

ChatGPT

01In ChatGPT on the web, go to Plugins, select the plus button, then Add custom MCP server.
02Name it superagnt, enter the server URL under Connection, and set Authentication to OAuth. The server URL is https://mcp.superagnt.com/mcp.
03Tick "I understand and want to continue", then select Create as a plugin. Sign in to superagnt_ when asked. No token needed.
04Install the plugin, then start a new chat, type @ and select superagnt.
05Ask: “Make a table for our team's decisions and save these: <decisions>”

ChatGPT takes custom MCP servers on the web, not on mobile, and only Business and Enterprise/Edu workspaces get write actions. On Pro, connected servers are read and fetch only, so saving rows needs one of those plans in ChatGPT (OpenAI).

Codex

01In a terminal, run codex mcp add superagnt --url https://mcp.superagnt.com/mcp, then codex mcp login superagnt. The login opens a browser to approve.
02Start a new Codex session.
03Ask: “Make a table for our team's decisions and save these: <decisions>”

04// memory or a database

When to use which

You want Claude to	Use	Why
Know your role, your tone, how you like answers	Claude's memory	It picks this up as you chat, with nothing to set up
Pick up a project where you left it	Claude's memory, inside a project	Each project keeps its own memory
Keep a list you'll add to for months: leads, decisions, content ideas	A superagnt_ table	Rows come back exactly, and you can count, sort and filter them
Share data with ChatGPT, Codex or a scheduled agent	A superagnt_ table	Every client and agent on the workspace reads the same database

05// how much it holds

How much the database holds

Each plan includes a workspace database. Tables: Starter 10, Growth 50 and Scale 250. Row allowance, across every table: Starter 25,000, Growth 250,000 and Scale 2,500,000. See pricing for the plans.