# MandateShield **Canonical:** https://apis.io/providers/mandateshield-com/ **Website:** https://mandateshield.com/ **APIs profiled:** 4 MandateShield is a fail-closed authority-verification and execution-evidence boundary for autonomous AI-agent purchases, operated from Bern, Switzerland. It binds a signed purchase approval (AP2, TAP, UCP, x402, MPP, ACP or a custom envelope) to a registered mandate and a pinned issuer key, reserves cumulative budget atomically, consumes the reservation into a one-use provider-bound execution permit for an exact Stripe PaymentIntent or x402 request, and then checks the provider outcome without trusting the caller before issuing a signed terminal receipt. The hosted service never moves money or holds payment credentials. It publishes an OpenAPI 3.1 contract with 26 operations, a remote MCP server whose tools/list answers anonymously, a signed A2A agent card, llms.txt, an SPDX SBOM, a JWKS, a 91-code reason registry, a 19-vector conformance suite and self-hosted JavaScript, Python and Go clients. ## Kin Score — 75.2 / 100 (exemplar) Scored 2026-10-09 under rubric 0.23.0. Trend: flat (+0.5 from 74.7). | Facet | Score | |---|---| | Discoverability | 80.0 | | Contract Quality | 53.4 | | Contract Governance | 18.2 | | Operational Transparency | 63.2 | | Developer Ergonomics | 80.4 | | Access Clarity | 84.2 | Regulatory layer — **Payments**: 50.0 (matched via tags). ## Agent readiness — 40.3 (agent-native) | Dimension | Value | |---|---| | Spec Presence | yes | | Agentic Access | derived | | Reversibility Documented | no | | MCP Server | documented | | Auth Clarity | bearer | | Idempotency | verified | | Error Semantics | verified | | OpenAPI Examples | partial | | Rate Limit Signal | documented | | Event Surface Described | no | | Agent Skills | derived | | Well Known Catalog | no | | Consent Identity | no | | Agent Card | flavored | | Dry Run Mode | no | | Delegated Identity | no | | Protected Resource Metadata | no | | Dynamic Client Registration | no | | Agentic Commerce | no | ## APIs (22) - **MandateShield MCP Server** — Hosted, stateless Streamable HTTP MCP server at https://mandateshield.com/api/mcp. Anonymous initialize and tools/list; three tools (check_ai_payment_authority, normalize_agent_... - **MandateShield Payment Authority Agent (A2A)** — JSONRPC A2A interface at https://mandateshield.com/a2a serving A2A 1.0 and 0.3 profiles (selected by the A2A-Version header) with three skills: project agent payment fields, ana... - **MandateShield Account execution control API** — Hosting-authenticated account-owner emergency control. The interlock governs new MandateShield-mediated strict operations inside one account regardless of provider profile. It c... - **MandateShield Advisory intelligence API** — Privacy-thresholded context. Signals can require REVIEW but do not automatically block. - **MandateShield Analysis only API** — Policy diagnostics that always return enforcement_authorized=false. - **MandateShield Deployment activation proofs API** — Signed, account-bound records of an externally bound subject reaching one server-observed strict live reservation and one fresh credential-bound permit redemption. The proof flo... - **MandateShield Deployment Proofs API** — The Deployment Proofs API from MandateShield — 3 operation(s) for deployment proofs. - **MandateShield Execution Authorizations API** — The Execution Authorizations API from MandateShield — 1 operation(s) for execution authorizations. - **MandateShield Execution evidence API** — Provider-bound permit redemption, caller observation intake, Stripe API or x402 chain reconciliation that does not trust the caller assertion, autonomous terminal authorization ... - **MandateShield Execution Permits API** — The Execution Permits API from MandateShield — 2 operation(s) for execution permits. - **MandateShield Execution Receipts API** — The Execution Receipts API from MandateShield — 1 operation(s) for execution receipts. - **MandateShield Normalize API** — The Normalize API from MandateShield — 1 operation(s) for normalize. - **MandateShield Preflight API** — The Preflight API from MandateShield — 1 operation(s) for preflight. - **MandateShield Production authority API** — Challenge, strict verification, and processor transitions. The documented eight-field invariant establishes only a RESERVED, consumable authorization; a trusted gateway must obt... - **MandateShield Proof Network API** — The Proof Network API from MandateShield — 5 operation(s) for proof network. - **MandateShield Protocol adapters API** — Stateless documented-field projection for supported agent-payment inputs. X402 and MPP require a canonical payee identity whose rail-specific fields exactly match the source. A ... - **MandateShield Provider Submissions API** — The Provider Submissions API from MandateShield — 1 operation(s) for provider submissions. - **MandateShield Public proof network API** — Signed externally bound self-reports. MandateShield verifies domain or exact GitHub-commit binding and claimant-report integrity; it does not run the claimant implementation, in... - **MandateShield Public sandbox API** — Zero-account, test-only and request-local lifecycle simulation. It uses ephemeral keys, contacts no external provider or independent organization, retains no lifecycle state and... - **MandateShield Receipts API** — Public-key signed-receipt verification and privacy-safe issuance lookup. - **MandateShield Sandbox API** — The Sandbox API from MandateShield — 1 operation(s) for sandbox. - **MandateShield Verify API** — The Verify API from MandateShield — 1 operation(s) for verify. ## MCP servers (3) - **MandateShield MCP Server** — One hosted, stateless Streamable HTTP MCP server at https://mandateshield.com/api/mcp. Anonymous initialize and tools/list both succeed (probed 2026-09-19); three tools are expo... - **MCP endpoint, no-auth plugin profile** - **MCP registry server.json** ## Agentic access (1) - **Mandateshield Com Agentic Access** — 43 operations · 17 acting · 1 human-in-the-loop ## Security (4) - **Mandateshield Com Authentication** — apiKey/http · 2 schemes - **Mandateshield Com Domain Security** — TLSv1.3 · HSTS · DMARC - **Mandateshield Com Vulnerability Disclosure** — security.txt · contact published - **Mandateshield Com Trust Center** — trust center published ## Plans (1) - **Mandateshield Com Plans Pricing** ## Tags Company, Payments, Agentic Commerce, AI Agents, Payment Authorization, Fraud Prevention, Cryptographic Verification, MCP, A2A, x402, AP2, Stripe, Agent-Native, Switzerland --- Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/mandateshield-com/). Scores are computed from the provider's own public artifacts under a published rubric.