Consensus for AI agents
Consensus searches scientific research papers for evidence-based answers, letting an agent ground a claim in peer-reviewed studies rather than general web content. 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 Consensus.
Search research on a topic
Search scientific papers on a topic to ground a claim in peer-reviewed evidence.
Check evidence for a claim
Search for studies that support or contradict a specific claim before citing it.
Gather sources for a summary
Pull several relevant papers on a topic before writing a research summary.
1 Consensus actions, graded by risk.
Every Consensus action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
1Look things up. Allowed by default.
- consensus.search_papersSearch Research Papers
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.
Consensus in three steps.
- 01Your users connect ConsensusThey add their Consensus 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 Consensus through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('consensus', { 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 Consensus connects.
Users add their Consensus 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 Consensus 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 Consensus from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Consensus and Arc0, answered.
Can I use Consensus with Claude, ChatGPT or Cursor?
Yes. Connect Consensus 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 Consensus actions you allow.
How do users connect Consensus?
Users add their Consensus 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 Consensus actions can my agent take?
1 in total: 1 read, 0 write and 0 destructive. Your policies decide which of them each agent may call.
Can I make my agent read-only in Consensus?
Yes. Allow read actions and deny writes in the Consensus policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Consensus too?
Yes. The same Consensus connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Consensus into your agent.
Your users connect Consensus once, under your brand. Your agent gets 1 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan