FutureAGI for AI agents
FutureAGI provides evaluation, observability, and annotation tools for building and improving AI systems before they ship, used by AI teams testing model quality. 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 FutureAGI.
Check an evaluation task's results
List eval tasks to see how a model version scored against a test set.
Review an annotation queue
List an annotation queue to see which items still need a human label.
Read-only dataset browsing
List datasets as a read-only step before deciding which one to run a new evaluation against.
9 FutureAGI actions, graded by risk.
Every FutureAGI action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
9Look things up. Allowed by default.
- futureagi.list_datasetsList Datasets
- futureagi.list_personasList Personas
- futureagi.list_run_testsList Run Tests
- futureagi.list_scenariosList Scenarios
- futureagi.list_eval_tasksList Eval Tasks
- futureagi.list_agent_definitionsList Agent Definitions
- futureagi.list_annotation_labelsList Annotation Labels
- futureagi.list_annotation_queuesList Annotation Queues
- futureagi.list_custom_eval_configsList Custom Eval Configs
write
0Create and change things. Allow, or ask the user first.
- No write actions.
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
FutureAGI in three steps.
- 01Your users connect FutureAGIThey add their FutureAGI api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “updates” can wait for the user to approve.
- 03Any agent can actYour agent calls FutureAGI through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('futureagi', { read: 'allow', write: 'ask', destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How FutureAGI connects.
Users add their FutureAGI 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 FutureAGI 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 FutureAGI from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
FutureAGI and Arc0, answered.
Can I use FutureAGI with Claude, ChatGPT or Cursor?
Yes. Connect FutureAGI 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 FutureAGI actions you allow.
How do users connect FutureAGI?
Users add their FutureAGI 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 FutureAGI actions can my agent take?
9 in total: 9 read, 0 write and 0 destructive. Your policies decide which of them each agent may call.
Can I make my agent read-only in FutureAGI?
Yes. Allow read actions and deny writes in the FutureAGI policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call FutureAGI too?
Yes. The same FutureAGI connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug FutureAGI into your agent.
Your users connect FutureAGI once, under your brand. Your agent gets 9 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan