All pages

What is KYA

• Was this AI agent allowed to do that?Was this AI agent allowed to do that? KYA verifies mandate, decision, receipt and timestamp independently and locally in your browser. Open, free, no account.
• Verify evidencePaste a mandate, a decision, a receipt chain or an RFC 3161 timestamp. Ten checks, locally in your browser, with a reproducible report.
• What is Know Your Agent?Know Your Agent checks whether an AI agent was authorized for a specific action: identity, principal, scope and proof. Definition, KYC comparison, building blocks.

Method

• The KYA methodHow KYA verifies an AI agent's authority: four questions, four verification areas, ten deterministic checks, three verdicts and five levels.
• Verification modelThe deterministic verification model of KYA: ten checks, six statuses, three verdicts, key provenance, time basis and all reason codes.
• Assurance levels A0 to A4The five cumulative levels of the assurance scale: from unverified through integrity, authority and provenance to time-anchored evidence. With exact rules.
• Error referenceAll 59 reason codes of the KYA engine with cause and fix, ordered by the ten checks. For developers and verifying parties.
• Error reference: Format and schemaReason codes in the “Format and schema” check: 6 codes with cause and fix, including format_unknown and format_recognized_not_supported.
• Error reference: Cryptographic integrityReason codes in the “Cryptographic integrity” check: 13 codes with cause and fix, including signature_invalid and no_key.
• Error reference: IssuerReason codes in the “Issuer” check: 3 codes with cause and fix, including issuer_key_not_trusted and kid_issuer_mismatch.
• Error reference: Agent and principalReason codes in the “Agent and principal” check: 5 codes with cause and fix, including agent_mismatch and principal_missing.
• Error reference: Mandate validityReason codes in the “Mandate validity” check: 3 codes with cause and fix, including expired and not_yet_valid.
• Error reference: Scope of the actionReason codes in the “Scope of the action” check: 7 codes with cause and fix, including action_not_in_scope and amount_exceeds_limit.
• Error reference: Delegation chainReason codes in the “Delegation chain” check: 4 codes with cause and fix, including no_delegation and delegation_broken.
• Error reference: RevocationReason codes in the “Revocation” check: 7 codes with cause and fix, including revoked and suspended.
• Error reference: Execution receiptReason codes in the “Execution receipt” check: 4 codes with cause and fix, including receipt_decision_mismatch and receipt_mandate_mismatch.
• Error reference: TimestampReason codes in the “Timestamp” check: 6 codes with cause and fix, including tsa_malformed and tsa_imprint_mismatch.

Standard

• The open standardKYA's specification, engine and test vectors are open. What is licensed, how versions work and how to measure an implementation.
• Know Your Agent approaches comparedA neutral comparison of Know Your Agent approaches from Skyfire, Akamai, Visa, Mastercard, Google and FIDO – and what KYA verifies. As of October 2026.
• ReleasesAll KYA releases with dates: new formats, checks and tools. The policy version kya-policy/1.0 remains stable; the engine follows Semantic Versioning.
• KYA verification profile 1.0The open KYA verification profile: how agent evidence is verified – claims profile, receipt chain kya-chain/1, evidence pack kya-pack/1, timestamps and report.
• Conformance and test vectors62 public test vectors with real signatures, including Verifiable Intent and Web Bot Auth. How to check your implementation and get a format added.
• Standards around agent authorityHow SD-JWT VC, status lists, RFC 3161, Web Bot Auth, AP2, Verifiable Intent and FIDO fit together – and where KYA comes in. As of October 2026.

Tools

• ToolsFree tools for agent evidence: decode SD-JWT and JWS, build the RFC 9421 signature base, read status lists and timestamps, locally in the browser.
• SD-JWT decoderDecode SD-JWT and SD-JWT VC locally: header, claims, disclosures with digest matching and the key binding JWT, without the token leaving the browser.
• JWS and JWT decoderDecode JWT and JWS locally: header, payload, times and signature length, without the token going to a server. For decisions, receipts and status lists.
• RFC 9421 signature baseBuild the RFC 9421 signature base from a signed HTTP request and compare it with your own implementation, locally in the browser, with the IETF example.
• Status list decoderDecode an IETF Token Status List locally: bits per entry, number of entries, revoked entries and the status at a given index.
• Timestamp inspector (RFC 3161)Read an RFC 3161 timestamp locally: time, hash algorithm, hash of the data, serial number, timestamp authority and embedded certificates.

Enterprise

• KYA for enterprisesHow enterprises use KYA: in the browser, in the CI pipeline or as a self-hosted copy in their own network. No data leaves, no license fees.
• Trust and securityHow KYA implements security and integrity: static architecture, strict security headers, release manifest with SHA-256, accessibility and vulnerability reporting.
• For developersInstall the KYA engine as an npm package and use it in code and CI pipelines: API, options, verification report, JSON schemas, command line and versioning.
• Financial servicesHow financial institutions verify whether an AI agent was authorized for a payment, order or claim: mandate, scope, revocation and receipt, recomputable.
• Commerce and paymentsHow merchants and payment providers verify whether an AI agent was allowed to buy on its customer's behalf: Verifiable Intent, Web Bot Auth, mandate and receipt.
• Public sectorHow public administration, courts and public enterprises prove on whose behalf an AI agent acted: mandate, delegation, receipt and an independent timestamp.
• AI agents in insurance: claims, policies, evidenceAI agents report claims, compare policies and conclude contracts on someone's behalf. What insurers and intermediaries must verify and how evidence helps.
• AI agents in travel and mobility: booking with evidenceAI agents book trains, flights and hotels, rebook during disruptions and manage mobility passes. What providers must consider to admit agents safely.
• AI agents in procurement: ordering within limitsAI agents order supplies, negotiate terms and manage suppliers. How companies transfer authority rules, signing rights and evidence to agents.
• AI agents in healthcare: administration with evidenceAI agents book doctor's appointments, submit invoices to health insurers and pass on requests. What practices and insurers verify and how data stays protected.
• AI agents in real estate: renting, managing, evidenceAI agents apply for apartments, book viewings and commission tradespeople. What property managers verify and where tenancy law requires a signature.
• AI agents in logistics: booking, customs, safeguardsAI agents book freight capacity, track shipments and prepare customs declarations. Which actions need limits and how representation in customs is proven.
• AI agents in telecommunications: contracts and SIM swapsAI agents switch plans, cancel contracts and request new SIM cards. Which actions telecom providers admit and why SIM swapping needs protection.

Knowledge

• QuestionsAnswers to common questions: what KYA checks, what happens to your data, what the levels mean and why a result is indeterminate.
• KYA, KYC, KYB: why agents need a check of their ownKnow Your Customer checks people, Know Your Business checks companies. Why AI agents need a third check that takes place on every action.
• Identity is not authority: what IAM leaves open for agentsAuthentication and access rights say who acts and what they may access. Why, for AI agents, that does not answer whether a specific action was allowed.
• What an auditor needs from an agent action: 12 questionsA checklist for audit, compliance and counterparties: the twelve questions evidence of an agent action must answer to be independently verifiable.
• Timestamps for agent receipts: plain, anchored, qualifiedWhat an RFC 3161 timestamp proves, why the trust anchor decides and how qualified timestamps under ZertES and eIDAS differ.
• Swiss e-ID and agent mandates: status October 2026The launch of the Swiss e-ID has been postponed, the trust infrastructure is targeted for the first half of 2027. What this means for mandates to AI agents.
• AI agent liability: who pays when an agent buys the wrong thing?Who bears the consequences when an AI agent orders the wrong item or exceeds its limits? An overview of the legal position in Switzerland, Germany and Austria.
• Proving authority: AI agents under Swiss, German and Austrian lawWhy proof of authority is decisive for AI agents, what verifiable evidence must contain and how it works under Swiss, German and Austrian law.
• Agentic commerce explained: how AI agents shopHow AI agents search, compare and pay, which approaches payment networks offer for this, and what merchants should check before accepting an agent purchase.
• Agent protocols compared: VI, AP2, TAP and Web Bot AuthVerifiable Intent, AP2, Trusted Agent Protocol, Web Bot Auth and Agentic Commerce Protocol compared: purpose, format, cryptography, openness, verifiability.
• Securing AI agents in MCP: mandates instead of API keysHow companies secure tool calls by AI agents through the Model Context Protocol: risks, authorization, decisions before execution and verifiable receipts.
• Prompt injection and authority: verification outside the modelWhy prompt injection cannot be fully prevented in AI agents and why checking authority must happen outside the language model.
• EU AI Act and AI agents: what applies to agentsWhat the European regulation on artificial intelligence means for AI agents: roles, transparency, record-keeping, deadlines and the situation in Switzerland.
• Adopting AI agents safely: a checklist including data protectionA practical checklist for companies adopting AI agents: authority, control, evidence, data protection under Swiss law and the GDPR, operations and contracts.
• Know Your Agent in Switzerland: verifying AI agentsWhat Know Your Agent means for Swiss companies: agency under the Code of Obligations, data protection, the e-ID with swiyu and a neutral verifier based in Zug.
• Know Your Agent in Germany: verifying AI agentsWhat Know Your Agent means for German companies: computer declarations and agency under the BGB, GDPR, the EUDI wallet, PSD2 and independent verification.
• Know Your Agent in Austria: verifying AI agentsWhat Know Your Agent means for Austrian companies: authorization under the ABGB, GDPR, ID Austria, the EUDI wallet and independent verification.
• Verify agent evidence in 5 stepsStep by step: verify evidence from an AI agent with KYA, provide the verifying party's trust, read verdict and level, and keep the certificate.
• Verify a signed HTTP request (RFC 9421, with IETF example)Guide: verify a signed HTTP request following RFC 9421 with KYA, using the official IETF example from Appendix B.2.6, including a counter-check with a changed path.
• Verify a timestamp (RFC 3161, with FreeTSA)Guide: create a timestamp following RFC 3161 with OpenSSL and FreeTSA, cross-check it locally and verify it in KYA with the timestamp anchor.
• Questions on law and liability for AI agentsAnswers to common legal questions about AI agents: liability, contract formation, authority, proof, data protection and regulation in CH, DE and AT.
• Questions on technology and formats in KYAAnswers to technical questions about KYA: formats, local verification, verdicts and levels, keys, revocation, timestamps, package and test vectors.
• State of the standards for AI agentsWhere the approaches for AI agents stand today, from Verifiable Intent and AP2 to Web Bot Auth, e-ID and the EUDI wallet, with chronology and KYA support.
• Agent mandate or OAuth token? The differenceOAuth governs an agent's access to an interface, an agent mandate the individual action on someone's behalf. Differences, use and interplay at a glance.
• Agent mandate or API key? Which is saferMany AI agents work with API keys. Why a key acts like blanket authority and what an agent mandate with limits, revocation and evidence does differently.
• SD-JWT VC or W3C Verifiable Credentials? ComparedTwo formats for digital credentials: the IETF's SD-JWT VC and the W3C Verifiable Credentials 2.0 data model. Structure, disclosure, adoption and fit for mandates.
• Web Bot Auth or bot detection? Recognizing agentsBot detection estimates from behavior, Web Bot Auth proves the origin of a request with a signature. What each achieves and why shops need both.
• RFC 3161 timestamp or blockchain anchoring?Two ways to prove a point in time for agent evidence independently: a timestamp from a timestamp authority following RFC 3161 or anchoring in a public blockchain.
• AI agent pays invoices: automating safelyAn AI agent checks and pays supplier invoices. How companies secure this with a mandate, a decision before payment, approval thresholds and verifiable receipts.
• AI agent books business travel: limits and evidenceAn AI agent books train, flight and hotel according to the travel policy. How mandate, delegation, decisions and receipts make bookings traceable.
• AI agent manages subscriptions: renewing and cancelingAn AI agent renews, switches and cancels subscriptions on a customer's behalf. Which limits should apply and how providers can verify actions by agents.
• AI agent in customer service: securing refundsAn AI agent handles complaints and books credits or refunds. How limits, decisions and receipts prevent abuse and mistakes in customer service.
• Use the engine in Node.js: verify evidence in the backendIntegrate the KYA engine in Node.js or TypeScript: install, provide trust, call verify(), evaluate verdict and checks. With working code.
• Verify evidence in the CI pipeline with kya-verifyVerify agent evidence automatically in GitHub Actions or any other CI pipeline: the kya-verify command line, exit codes, trust in the repository, report as JSON.
• Publish keys as JWKS: a guide for issuersAs an issuer of agent evidence, create a key pair, publish the public key as JWKS and rotate keys without invalidating older evidence.
• Publish and maintain a status list: revocation for mandatesCreate, compress, sign and publish a status list following the IETF Token Status List so mandates can be revoked before they expire. With working code.
• Questions from merchants about AI agents: 13 answersWhat online shops and merchants need to know about shopping AI agents: recognizing, admitting, checking orders, payments, returns, liability and first steps.
• Questions from developers about KYA: 13 answersAnswers for developers: integrating the engine, formats, keys, status lists, exit codes, test vectors, determinism, license and performance.
• Guardrails or authorization? What secures AI agentsGuardrails filter what a model says; authorization decides what an agent may do. Why actions with consequences need a deterministic check.
• MCP or A2A? Two protocols for AI agents comparedMCP connects agents with tools, A2A connects agents with each other. Purpose, structure, security and how both protocols work together in one architecture.
• Decision before or log after? Controlling agentsMany companies log what their AI agents do. Why a log prevents no damage and what a signed decision before execution changes.
• Human approval or a mandate with limits?Should a person approve every action of an AI agent, or should a mandate set the limits? Strengths, weaknesses and the combination that works in practice.
• Trusted Agent Protocol or Web Bot Auth? ComparedVisa's Trusted Agent Protocol and Web Bot Auth sign agent requests following RFC 9421. What they share, how they differ and what merchants take from it.
• Agent mandate or written authorization on paper?A written authorization on paper is made for people, an agent mandate for software. What both achieve, where the law requires form and how companies combine them.
• AI agent shops online: purchase intent, limits, evidenceAn AI agent shops online on a customer's behalf. How purchase intent, limits and evidence make the purchase safe and traceable for customer and merchant.
• AI agent files applications with authorities: proving representationAn AI agent fills in forms and files applications and notifications with authorities. How representation, identity and evidence work together in e-government.
• AI agent books appointments: limits, cancellations, privacyAn AI agent arranges, reschedules and cancels appointments with practices, tradespeople and service providers. Which limits apply and how providers verify bookings.
• AI agent creates sales offers: binding within limitsAn AI agent creates and sends offers on a company's behalf. Why binding offers need clear limits and how customers verify the agent's authority.
• AI agent manages cloud resources: limits in operationsAn AI agent scales servers, adjusts configurations and cuts cloud costs. How mandate, approvals and receipts prevent outages and expensive mistakes.
• AI agent runs code deployments: safely into productionCoding agents write code, open pull requests and deploy. How mandate, protected branches, approvals and receipts secure deliveries by agents.
• KnowledgeArticles, glossary and questions on KYA, identity and authority, evidence for auditors, timestamps and the Swiss e-ID – with sources.
• GlossaryDefinitions of agent mandate, decision token, receipt chain, status list, key binding, trust anchor and further terms around Know Your Agent.

About

• About KYAWho operates KYA, which principles apply, how the relationship with the reference implementation Mandact is governed and how to use the brand.
• ContactQuestions, new formats, security reports or press: how to reach the team behind KYA. Your message goes directly from your own email app to us.
• Legal noticeLegal notice of knowyouragent.ch: Innopulse Consulting GmbH, Gotthardstrasse 30, 6300 Zug, UID CHE-219.727.921. Contact: info@innopulse.io.
• Privacy policyPrivacy policy of knowyouragent.ch: verification locally in the browser, no cookies, no tracking, contact through your own email app. Your rights under Swiss law.
• Leutrim MiftarajLeutrim Miftaraj is the founder and managing director of Innopulse Consulting GmbH in Zug, Switzerland, and designed KYA's method, specification and engine.
• Reference implementation: MandactMandact issues agent mandates, decides before every execution and produces chained receipts. Its role as reference implementation and how separation works.

Glossary

Know Your Agent (KYA) · Agent mandate · Principal · Agent passport · Decision token · Execution receipt · Receipt chain · Evidence pack · Verifying party · Verifier · Fail-closed · SD-JWT VC · Key binding · Status list · Timestamp (RFC 3161) · Trust anchor · Delegation chain · Blanket authority · AI agent · Agentic commerce · Authority to act · Verifiable Intent · Agent Payments Protocol (AP2) · Web Bot Auth · HTTP Message Signatures (RFC 9421) · Trusted Agent Protocol · Selective disclosure · Decentralized Identifier (DID) · OpenID for Verifiable Presentations (OID4VP) · Qualified timestamp · Prompt injection · Least privilege · Model Context Protocol (MCP) · Swiss e-ID (swiyu) · Rich Authorization Requests (RFC 9396) · OAuth Token Exchange (RFC 8693) · DPoP (RFC 9449) · OAuth 2.1 · Mutual TLS (RFC 8705) · Agent2Agent protocol (A2A) · Agent Card · Agentic Commerce Protocol (ACP) · Confused deputy · Replay attack · EUDI wallet · eIDAS 2.0 · ID Austria · Qualified electronic signature · W3C Verifiable Credentials 2.0 · Strong customer authentication · Network tokenization · Human in the loop · Policy decision point (PDP) · Zero trust · JWKS (JSON Web Key Set) · X.509 certificate · Canonicalization (JCS) · Workload identity (SPIFFE) · Passkeys · Know Your Business (KYB) · Tool poisoning · Guardrails · Payment initiation service · Electronic seal · OpenID Federation