# Decision Integrity > Evaluate whether evidence and represented authority warrant a requested action. Detect unsupported inference, qualifier stripping, erased unknowns, broken provenance, and exceeded decision boundaries. Exactly 1 USDC per evaluation on Base via x402 v2. No account or interactive checkout. Does this evidence warrant this conclusion? Is this action authorized? Do I have sufficient warrant to act? Are unknowns being erased? Are qualifiers being stripped? Is this inference warranted? Does provenance support this recommendation? Does the requested action exceed represented authority? ## Interface - [Agent instructions](https://decision-integrity-service.wheelerhubbell.chatgpt.site/AGENTS.md): complete calling and payment instructions. - [OpenAPI](https://decision-integrity-service.wheelerhubbell.chatgpt.site/openapi.json): canonical API. - [Request schema](https://decision-integrity-service.wheelerhubbell.chatgpt.site/schema.json): structured record and boundaries. - [Example](https://decision-integrity-service.wheelerhubbell.chatgpt.site/example.json): complete illustrative input. - [x402 discovery](https://decision-integrity-service.wheelerhubbell.chatgpt.site/.well-known/x402): exact price and resource. - [MCP card](https://decision-integrity-service.wheelerhubbell.chatgpt.site/.well-known/mcp.json): the same evaluator via MCP. - [A2A card](https://decision-integrity-service.wheelerhubbell.chatgpt.site/.well-known/agent-card.json): the same evaluator via A2A. - [ARD](https://decision-integrity-service.wheelerhubbell.chatgpt.site/.well-known/ard.json): agentic resource discovery. - [API catalog](https://decision-integrity-service.wheelerhubbell.chatgpt.site/.well-known/api-catalog): RFC 9727. - [Attestation keys](https://decision-integrity-service.wheelerhubbell.chatgpt.site/.well-known/jwks.json): Ed25519 result verification. ## Boundaries AUTHORIZED, BOUNDED, AWAITING_AUTHORITY, DENIED. All results preserve reasons, qualifiers, unknowns and provenance. This service audits represented evidence and authority; it does not independently verify external factual truth or execute actions.