core

Firewall an LLM response — verify claims, detect hallucinations, and enforce output policy.

Cost: 2 credit(s). Free tier: 10 calls per IP (no auth).

post/v1/verify

Request

  • Base URL: https://api.agent-toolbox.ai
  • URL: https://api.agent-toolbox.ai/v1/verify
  • Auth: one of:
    • HTTP bearer
    • none

Request body

outputType'code' | 'natural_language' | 'agent_action' | 'factual_claim' required
llmResponsestring required
language'python' | 'javascript' | 'typescript' | 'rust' | 'go'
enforcementMode'block' | 'flag' | 'audit'
timeoutMsinteger

Response

Success.

verdict'PASS' | 'FLAG' | 'BLOCK' required
overallScorenumber required
outputType'code' | 'natural_language' | 'agent_action' | 'factual_claim' required
enforcementMode'block' | 'flag' | 'audit' required
latencyMsnumber required

Server compute time in ms.

certificatestring required

Tamper-evident SHA-256 certificate.

Example response

{
  "certificate": "sha256:5bbf2510c504a1c1"
}

Changes

No recorded changes to this operation across all 1 revision of this API.