Algorand
Check an agent before you pay it.
One call. A clear answer. Paid in USDC on Algorand.
The live numbers could not be read right now. The rail itself is unaffected.
Try it
Try
The oracle could not be reached just now. Try again in a few seconds.
Free preview. Your agent gets the signed verdict for $0.05.
Add it to your agent
Your agent pays from its own account. Anything above its limit is refused before it signs.
Claude, Cursor, any MCP client
One command
claude mcp add a-identity-trust \ -e A_IDENTITY_ALGORAND_MNEMONIC="your 25 words" \ -- npx -y @a-identity/trust-mcp
TypeScript
npm install @a-identity/trust-guard algosdk
import { TrustGuard } from '@a-identity/trust-guard'
import { algorandPayer } from '@a-identity/trust-guard/algorand'
const oracle = new TrustGuard({ rail: 'algorand', onPaymentRequired: algorandPayer({ mnemonic }) })
await oracle.guard(agentId) // throws if you should not payAsk one question, pay for one answer
Should I pay this agent?
risk_check
Is it who it says it is?
verify_agent
How has it behaved?
reputation_score
Everything, in one answer
agent_passport
A whole shortlist at once
agent_batch_audit