Retable for AI agents
Retable is a collaborative spreadsheet and database tool for organizing workspaces, projects, tables, and rows that teams can query and edit together. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Retable.
Search for a row
Search rows in a table to find the record that needs updating.
Insert new rows
Insert rows into a table after collecting new data from another source.
Confirm before bulk deleting
Get the rows first, then delete them only after confirming the right ones are selected.
11 Retable actions, graded by risk.
Every Retable action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
6Look things up. Allowed by default.
- retable.get_rowsGet Rows
- retable.get_tableGet Table
- retable.search_rowsSearch Rows
- retable.list_workspacesList Workspaces
- retable.list_project_tablesList Project Tables
- retable.list_workspace_projectsList Workspace Projects
write
3Create and change things. Allow, or ask the user first.
- retable.add_columnsAdd Columns
- retable.update_rowsUpdate Rows
- retable.insert_rowsInsert Rows
destructive
2Delete, cancel or archive. Ask first, or deny outright.
- retable.delete_rowsDelete Rows
- retable.delete_columnsDelete Columns
Retable in three steps.
- 01Your users connect RetableThey add their Retable api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “update Rows” can wait for the user, and “delete Rows” can be denied outright.
- 03Any agent can actYour agent calls Retable through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('retable', { read: 'allow', write: 'ask', // update_rows destructive: 'deny', // delete_rows }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Retable connects.
Users add their Retable api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Retable connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Retable from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraphMore databases and spreadsheets apps.
Retable and Arc0, answered.
Can I use Retable with Claude, ChatGPT or Cursor?
Yes. Connect Retable to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Retable actions you allow.
How do users connect Retable?
Users add their Retable api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Retable actions can my agent take?
11 in total: 6 read, 3 write and 2 destructive, such as “update Rows”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Retable?
Yes. Actions like “delete Rows” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call Retable too?
Yes. The same Retable connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Retable into your agent.
Your users connect Retable once, under your brand. Your agent gets 11 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan