Network Working Group H. S. Ganesan Internet-Draft 15 November 2026 Intended status: Standards Track Expires: 19 May 2027 Agent Authenticator: Hardware-Bound, Origin-Bound Authentication for AI Agents draft-ganesan-agent-authenticator-00 Abstract This document specifies the Agent Authenticator, a mechanism that lets autonomous software agents, such as AI coding agents, authenticate to services with properties comparable to passkeys: keys held in secure hardware, no shared secrets, binding to the service's origin that a manipulated agent cannot override, attested execution environments, and explicit human approval of each agent and of high- risk actions. The same protocol serves agents on personal computers and in cloud environments. It is a profile over existing mechanisms rather than a new identity system: it uses OAuth 2.0 Token Exchange, the Identity Assertion JSON Web Token (JWT) Authorization Grant, and Model Context Protocol (MCP) authorization, and it binds each agent action to a signed, tamper-evident record. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at https://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on 19 May 2027. Copyright Notice Copyright (c) 2026 IETF Trust and the persons identified as the document authors. All rights reserved. Ganesan Expires 19 May 2027 [Page 1] Internet-Draft Agent Authenticator November 2026 This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/ license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1. Goals and Non-Goals . . . . . . . . . . . . . . . . . . . 3 1.2. Motivation . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.1. Evidence from Real Incidents . . . . . . . . . . . . 4 1.2.2. Evidence from Credential Data . . . . . . . . . . . . 5 1.2.3. Gaps in Existing Standards . . . . . . . . . . . . . 5 1.3. Requirements Language . . . . . . . . . . . . . . . . . . 6 2. Terminology and Architecture . . . . . . . . . . . . . . . . 6 2.1. Abbreviations . . . . . . . . . . . . . . . . . . . . . . 8 3. Security Properties . . . . . . . . . . . . . . . . . . . . . 9 4. Keys and Assurance Levels . . . . . . . . . . . . . . . . . . 11 5. Data Formats . . . . . . . . . . . . . . . . . . . . . . . . 12 5.1. Mandate . . . . . . . . . . . . . . . . . . . . . . . . . 12 5.2. AgentData . . . . . . . . . . . . . . . . . . . . . . . . 13 5.3. Assertion . . . . . . . . . . . . . . . . . . . . . . . . 14 5.4. Attestation Statement . . . . . . . . . . . . . . . . . . 14 5.5. AgentData Claim Definitions . . . . . . . . . . . . . . . 14 6. Ceremonies . . . . . . . . . . . . . . . . . . . . . . . . . 15 6.1. Registration . . . . . . . . . . . . . . . . . . . . . . 15 6.2. Authentication . . . . . . . . . . . . . . . . . . . . . 16 6.3. Step-Up . . . . . . . . . . . . . . . . . . . . . . . . . 16 6.4. Sub-Delegation . . . . . . . . . . . . . . . . . . . . . 17 6.5. Revocation . . . . . . . . . . . . . . . . . . . . . . . 18 6.6. Relying Party Verification Procedure . . . . . . . . . . 18 7. Protocol Bindings . . . . . . . . . . . . . . . . . . . . . . 19 8. Action Records . . . . . . . . . . . . . . . . . . . . . . . 20 8.1. What Is Recorded . . . . . . . . . . . . . . . . . . . . 20 8.2. Record Generation . . . . . . . . . . . . . . . . . . . . 21 8.3. Tamper Evidence . . . . . . . . . . . . . . . . . . . . . 22 8.4. SIEM Delivery . . . . . . . . . . . . . . . . . . . . . . 23 9. Deployment Profiles . . . . . . . . . . . . . . . . . . . . . 23 10. Implementation Status . . . . . . . . . . . . . . . . . . . . 24 11. Security Considerations . . . . . . . . . . . . . . . . . . . 25 12. Privacy Considerations . . . . . . . . . . . . . . . . . . . 27 13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 27 13.1. Media Types . . . . . . . . . . . . . . . . . . . . . . 27 13.1.1. application/agent-mandate+jwt . . . . . . . . . . . 27 Ganesan Expires 19 May 2027 [Page 2] Internet-Draft Agent Authenticator November 2026 13.1.2. application/agent-assertion+jwt . . . . . . . . . . 28 13.2. OAuth URI . . . . . . . . . . . . . . . . . . . . . . . 28 13.3. JSON Web Token Claims . . . . . . . . . . . . . . . . . 28 13.4. Agent Authenticator Registry Group . . . . . . . . . . . 29 13.4.1. Agent Assurance Levels Registry . . . . . . . . . . 29 13.4.2. Agent Action Decision Values Registry . . . . . . . 30 14. References . . . . . . . . . . . . . . . . . . . . . . . . . 30 14.1. Normative References . . . . . . . . . . . . . . . . . . 30 14.2. Informative References . . . . . . . . . . . . . . . . . 31 Appendix A. Standardization Path . . . . . . . . . . . . . . . . 34 Appendix B. Related Work . . . . . . . . . . . . . . . . . . . . 34 Appendix C. Acknowledgments . . . . . . . . . . . . . . . . . . 34 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 35 1. Introduction Software agents built on large language models increasingly act on behalf of people when using services such as code hosts, package registries, model APIs, MCP servers and cloud platforms. This document specifies how an agent authenticates to those services with properties comparable to passkeys, without giving the agent any secret it could leak. It defines one new component, the Agent Authenticator, along with its data formats, ceremonies and bindings to existing OAuth, MCP, git and cloud mechanisms. 1.1. Goals and Non-Goals The mechanism is designed to meet the following goals: 1. The agent process does not hold private keys or any other secret that can be used to obtain tokens. It requests signatures from a separate authenticator. 2. A credential can be used only with the relying party for which it was created. The origin is taken from the network connection rather than from the agent's output, so an agent that has been manipulated through prompt injection cannot direct a credential to another site. 3. Agent keys are generated and held in secure hardware or in an attested cloud key service, and relying parties can learn how well a key is protected through a normalized assurance level. 4. A named person approves each agent credential, and each high-risk action, using a passkey. The person's device need not be the device on which the agent runs. Ganesan Expires 19 May 2027 [Page 3] Internet-Draft Agent Authenticator November 2026 5. Deployment does not require a new identity system. Enterprise identity providers remain the policy authority, and services accept agent credentials through existing OAuth grants. 6. Each action that the authenticator authorizes is recorded in a signed, tamper-evident log that can be exported to existing security monitoring systems. 7. The same protocol is used whether the agent runs on a personal computer or in a cloud environment; only the implementation of the authenticator differs. The following are out of scope: replacing OAuth, OpenID Connect or WebAuthn; agent-to-agent messaging; and assessing whether an agent's output is correct. 1.2. Motivation Coding agents today authenticate with the same bearer secrets humans use (PATs, API keys, npm tokens, cloud keys), stored where the model, its subprocesses and any malware can read them. The incidents below, all reported in 2025 or 2026, illustrate the resulting failures. 1.2.1. Evidence from Real Incidents +==============+===================+==============+================+ | Incident | What happened | Root cause | Property | | (date) | | | | +==============+===================+==============+================+ | GitHub MCP | A malicious issue | Broad PAT | P8 mandate | | prompt | in a public repo | covering | scopes repos | | injection | made an agent | every repo | and actions; | | (May 2025) | pull private repo | the user can | P5 step-up for | | | data and publish | reach | cross-repo | | | it in a public PR | | writes | +--------------+-------------------+--------------+----------------+ | Comment and | One PR title | Secrets | P2 model never | | Control (Apr | hijacked several | ambient in | holds secrets; | | 2026) | CI coding agents, | the agent's | P8 review | | | which leaked repo | environment | agents get no | | | secrets back | | write or | | | through GitHub | | secret scopes | | | itself | | | +--------------+-------------------+--------------+----------------+ | Six coding- | Branch names | Controls | AA enforces | | agent | stole tokens; | lived inside | policy outside | | exploits | repo settings | the agent | the agent; P9 | | (2025-2026) | files skipped | and could be | every agent | Ganesan Expires 19 May 2027 [Page 4] Internet-Draft Agent Authenticator November 2026 | | trust dialogs; | bypassed | registered and | | | sandboxes were | from repo | tied to a | | | escaped | content | human | +--------------+-------------------+--------------+----------------+ | Nx | Malicious package | Tokens and | P1, P2 nothing | | s1ngularity | versions | keys | to steal; AA | | (Aug 2025) | harvested 2,349 | readable on | signs only for | | | credentials from | disk by any | processes it | | | 1,079 developer | local | launched | | | machines, using | process | | | | local AI CLIs to | | | | | hunt for secrets | | | +--------------+-------------------+--------------+----------------+ | Shai-Hulud | Self-propagating | Long-lived | P7 no | | worm waves | worm steals npm | publish | exportable | | (Sep 2025 - | and GitHub tokens | tokens | credentials; | | Oct 2026) | and republishes | usable from | publish | | | itself into every | any machine | requires ha | | | package the | | | | | victim maintains | | | +--------------+-------------------+--------------+----------------+ Table 1 Sources: [INVARIANT], [CSA-CI], [VB-EXPLOITS], [NX-LEAK] and [SHAI-HULUD]. 1.2.2. Evidence from Credential Data * A study by Astrix Security [ASTRIX] of 5,205 open-source MCP servers found 88% need credentials, 53% use static API keys or PATs, only 8.5% use OAuth, and 79% pass keys through environment variables. * GitGuardian [GG-SPRAWL] found 28.65 million newly leaked secrets on public GitHub in 2025, an 81% increase in leaked secrets for AI services, and 24,008 secrets in MCP configuration files, 2,117 of them valid. 1.2.3. Gaps in Existing Standards +===============+=====================+============================+ | Standard | What it gives | What is missing for agents | +===============+=====================+============================+ | WebAuthn / | Phishing-resistant, | Needs a human gesture on | | passkeys | hardware-bound keys | every use; no programmatic | | | | enrollment | +---------------+---------------------+----------------------------+ Ganesan Expires 19 May 2027 [Page 5] Internet-Draft Agent Authenticator November 2026 | OAuth 2.1 / | Scoped, expiring | Bearer tokens by default; | | MCP | tokens; standard | no binding to a device or | | authorization | discovery | attested environment | +---------------+---------------------+----------------------------+ | ID-JAG / | IdP decides which | Based on the user's | | Cross App | apps an agent may | session; no proof of where | | Access | reach | the agent key lives or | | | | that a human approved it | +---------------+---------------------+----------------------------+ | Workload | Secretless identity | Cloud-centric; no human | | identity | for cloud workloads | approver; weak fit for | | (SPIFFE, | | laptops | | WIMSE, cloud | | | | federation) | | | +---------------+---------------------+----------------------------+ | Secret vaults | Keep secrets out of | Still bearer secrets | | and proxies | the model's context | underneath; no standard | | | | format, origin binding or | | | | attestation | +---------------+---------------------+----------------------------+ Table 2 The author is not aware of an existing standard that combines hardware-bound keys, origin binding that survives prompt injection, a human-signed scope, environment attestation, and a single protocol for local and cloud agents. 1.3. Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here. 2. Terminology and Architecture The architecture involves the roles listed below. Of these, only the Agent Authenticator is introduced by this document. Ganesan Expires 19 May 2027 [Page 6] Internet-Draft Agent Authenticator November 2026 +===============+==============================+===============+ | Role | What it is | Passkey | | | | analog | +===============+==============================+===============+ | Human | The accountable person who | User | | approver | owns or delegates to the | | | | agent | | +---------------+------------------------------+---------------+ | Approver | The human's passkey device | Authenticator | | authenticator | (phone, laptop, security | | | | key) | | +---------------+------------------------------+---------------+ | Agent | The model-driven process | None | | | (coding agent, its | (untrusted | | | subprocesses and tools) | caller) | +---------------+------------------------------+---------------+ | Agent runtime | The harness that launches | Browser / | | | the agent and owns its | client | | | network connections | | +---------------+------------------------------+---------------+ | Agent | Daemon that holds agent keys | Authenticator | | Authenticator | in secure hardware, enforces | | | (AA) | mandates, signs assertions | | | | and action records | | +---------------+------------------------------+---------------+ | Relying party | Service the agent calls: | Relying party | | (RP) | code host, package registry, | | | | model API, MCP server, cloud | | +---------------+------------------------------+---------------+ | Identity | Enterprise policy authority, | None | | provider | such as a single sign-on | | | (IdP) | provider | | +---------------+------------------------------+---------------+ | Evidence | Local sealed log, SIEM and a | None | | pipeline | SCITT transparency service | | +---------------+------------------------------+---------------+ Table 3 Ganesan Expires 19 May 2027 [Page 7] Internet-Draft Agent Authenticator November 2026 +--------------------+ | Enterprise IdP | | policy, revocation | +---------+----------+ | token exchange | (ID-JAG) +-----------+ call +---------------+ assert. +--------------------+ | Agent |------->| Agent runtime |-------->| Relying parties | | (no keys) | | sets origin | | GitHub, npm, MCP | +-----------+ +---+-------^---+ +--------------------+ sign | | signed request v | assertion +-----------+ approval +---------------+ revocation (SSF/CAEP) | Human |----------->| Agent |<------------- from IdP | (passkey) | | Authenticator | +-----------+ | keys in HW | +---------------+ Figure 1: Architecture overview The agent can request a signature but never receives a key. The runtime supplies the origin of each request from its own connection state, and only the Agent Authenticator holds keys. Where an IdP is present, it remains the policy authority and can revoke an agent's access. This document uses the following terms: Agent credential: An RP-scoped key pair created and held by the AA for one agent instance. Mandate: A human-signed statement of what an agent credential may do, where, and until when. Environment attestation: Hardware evidence about where the AA's keys live and what runs there. Assurance level (AL): A normalized rating of environment attestation (Section 4). Action record: A signed, hash-chained record of one agent action, bound to the credential that authorized it. 2.1. Abbreviations AA: Agent Authenticator AL: assurance level CAEP: Continuous Access Evaluation Profile Ganesan Expires 19 May 2027 [Page 8] Internet-Draft Agent Authenticator November 2026 COSE: CBOR Object Signing and Encryption DPoP: Demonstrating Proof of Possession EAT: Entity Attestation Token ID-JAG: Identity Assertion JWT Authorization Grant IdP: identity provider KMS: key management service MCP: Model Context Protocol OCSF: Open Cybersecurity Schema Framework PAT: personal access token RAR: Rich Authorization Requests RATS: Remote ATtestation procedureS RP: relying party SCITT: Supply Chain Integrity, Transparency, and Trust SIEM: security information and event management SSF: Shared Signals Framework TEE: trusted execution environment TPM: Trusted Platform Module 3. Security Properties The following table compares the security properties of passkeys with those of this mechanism. Properties P8 to P10 have no counterpart in passkeys. +=======+================+====================+=====================+ | # | Property | Passkeys | This specification | +=======+================+====================+=====================+ | P1 | No shared | Public-key | Same; RPs store | | | secrets | challenge-response | only public keys | +-------+----------------+--------------------+---------------------+ | P2 | Key non- | Platform or | TPM, Secure | | | exportable | roaming | Enclave, | | | | authenticator | StrongBox, TEE or | | | | | attested cloud | | | | | KMS; model process | | | | | never sees keys | +-------+----------------+--------------------+---------------------+ | P3 | Origin binding | Browser sets | Runtime sets | | | | origin; key scoped | origin from the | | | | to RP ID | TLS connection; | | | | | key scoped to RP | | | | | ID; model output | | | | | can never choose | | | | | it | +-------+----------------+--------------------+---------------------+ | P4 | Attestation | Optional | Required | | | | authenticator | environment | | | | attestation | attestation, | Ganesan Expires 19 May 2027 [Page 9] Internet-Draft Agent Authenticator November 2026 | | | | normalized to | | | | | assurance levels | +-------+----------------+--------------------+---------------------+ | P5 | Presence / | UP and UV flags | MB flag (acting | | | verification | | within mandate) | | | | | and HA flag (human | | | | | approved this | | | | | action) | +-------+----------------+--------------------+---------------------+ | P6 | Clone | Signature counter | Signature counter | | | detection | | plus action-record | | | | | chain head in | | | | | every assertion | +-------+----------------+--------------------+---------------------+ | P7 | Sync | Passkeys may sync | Agent credentials | | | | | are never | | | | | synchronized; in | | | | | the cloud profile | | | | | they are destroyed | | | | | when the task ends | +-------+----------------+--------------------+---------------------+ | P8 | Scoped | None | Mandate: RP list, | | (new) | authority | | actions, | | | | | resources, expiry, | | | | | minimum assurance | +-------+----------------+--------------------+---------------------+ | P9 | Accountability | None | Every assertion | | (new) | | | names its human | | | | | approver and | | | | | mandate hash | +-------+----------------+--------------------+---------------------+ | P10 | Identity-bound | None | Every authorized | | (new) | evidence | | action yields a | | | | | signed, hash- | | | | | chained action | | | | | record | +-------+----------------+--------------------+---------------------+ Table 4 P3 is the property that most directly addresses prompt injection. If an agent is instructed to send a credential to a server controlled by an attacker, it cannot do so: no credential exists for that origin, and the agent has no way to set the origin itself. Ganesan Expires 19 May 2027 [Page 10] Internet-Draft Agent Authenticator November 2026 4. Keys and Assurance Levels The mechanism uses four keys. None of them is accessible to the agent process. +============+===============+=================+=================+ | Key | Held by | Created | Lifetime | +============+===============+=================+=================+ | Approver | Human's | Normal WebAuthn | Long-lived; | | key | passkey | registration | device-bound | | | authenticator | with the IdP or | passkey | | | | RP | RECOMMENDED for | | | | | approvals | +------------+---------------+-----------------+-----------------+ | AA | AA secure | AA install or | Life of the | | identity | hardware | cloud | device or | | key | | environment | environment | | | | boot | | +------------+---------------+-----------------+-----------------+ | Agent | AA secure | Per agent | Local: until | | credential | hardware | instance, per | revoked or | | key | | RP | mandate expiry. | | | | | Cloud: one task | +------------+---------------+-----------------+-----------------+ | Evidence | AA secure | With the AA | Rotated per | | signing | hardware, | identity key | epoch, forward- | | key | separate from | | secure | | | credential | | | | | keys | | | +------------+---------------+-----------------+-----------------+ Table 5 Verifiers MUST base policy decisions on the normalized assurance level defined below, rather than on vendor-specific attestation formats. Ganesan Expires 19 May 2027 [Page 11] Internet-Draft Agent Authenticator November 2026 +=======+===========================+===========================+ | Level | Meaning | Example evidence | +=======+===========================+===========================+ | AL3 | Keys in a TEE that also | AWS Nitro, Intel TDX, AMD | | | isolates the agent's | SEV-SNP, Arm CCA | | | code; runtime measured | attestation report | +-------+---------------------------+---------------------------+ | AL2 | Keys in dedicated secure | TPM 2.0 key attestation, | | | hardware; host software | Apple App Attest, Android | | | not isolated | StrongBox key attestation | +-------+---------------------------+---------------------------+ | AL1 | Keys in a platform | Android TEE-only | | | keystore without hardware | keystore, cloud instance | | | isolation, or cloud | identity document, SPIFFE | | | workload identity only | SVID | +-------+---------------------------+---------------------------+ | AL0 | Software key | None | +-------+---------------------------+---------------------------+ Table 6 The AA converts vendor evidence into an Entity Attestation Token [RFC9711] carrying al, platform, key-protection claims and a freshness nonce. A RATS [RFC9334] verifier appraises it; a separate EAT profile is expected to standardize the mapping. A passkey that is synchronized between devices authenticates the person but does not identify the device that was used. Approvals that require device binding (Section 6.3) MUST use a device-bound authenticator or carry a separate platform device attestation. 5. Data Formats All structures are JSON Web Signatures [RFC7515] for HTTP and OAuth bindings, with a COSE [RFC9052] encoding for constrained and SSH bindings. 5.1. Mandate The mandate is signed by the human approver. Ganesan Expires 19 May 2027 [Page 12] Internet-Draft Agent Authenticator November 2026 { "typ": "agent-mandate+jwt", "iss": "", "sub": "", "agent_cred": { "jkt": "" }, "aa": { "id": "", "min_al": "AL2" }, "rps": ["code.example", "registry.example", "mcp.example"], "authorization_details": [ { "type": "git", "actions": ["push"], "locations": ["code.example/acme/api"], "refs": ["agent/fix-*"] }, { "type": "mcp", "tools": ["jira.read", "jira.comment"] } ], "step_up": ["merge", "publish", "secrets.read", "delete"], "delegable": false, "exp": 1762480000 } The jkt member is a JWK Thumbprint [RFC7638]. The approver signs the mandate through a WebAuthn [WebAuthn] assertion whose challenge is SHA-256(mandate). The authorization_details member reuses [RFC9396] so IdPs and RPs evaluate it with existing policy engines. 5.2. AgentData AgentData is the signed core of every assertion, analogous to WebAuthn authenticatorData. { "rp_id": "code.example", "origin": "https://api.code.example", "challenge": "", "cred_id": "", "flags": { "mb": true, "ha": false, "ev": true }, "mandate_hash": "", "al": "AL2", "counter": 412, "evidence_head": "", "action": { "type": "git.push", "target": "acme/api@agent/fix-123" } } The origin and rp_id members are filled by the runtime from the live connection. The AA MUST reject any request where they do not match a credential's scope. Ganesan Expires 19 May 2027 [Page 13] Internet-Draft Agent Authenticator November 2026 5.3. Assertion The assertion is AgentData signed by the agent credential key. When ha is set, it also carries the approver's WebAuthn assertion over SHA-256(AgentData) for this specific action. 5.4. Attestation Statement Used at registration only: an EAT [RFC9711] containing the AA identity key, the agent credential public key, al, platform evidence (raw vendor statement, optional), the RP-provided nonce, and the environment measurement where AL3 applies. 5.5. AgentData Claim Definitions +===============+=========+=============+=======================+ | Claim | Type | Requirement | Meaning | +===============+=========+=============+=======================+ | rp_id | string | REQUIRED | RP ID the credential | | | | | is scoped to; set by | | | | | the runtime | +---------------+---------+-------------+-----------------------+ | origin | string | REQUIRED | Origin of the live | | | | | TLS connection; never | | | | | set by the agent | +---------------+---------+-------------+-----------------------+ | challenge | string | REQUIRED | Fresh nonce from the | | | | | RP, at least 128 bits | +---------------+---------+-------------+-----------------------+ | cred_id | string | REQUIRED | Identifier of the | | | | | agent credential | +---------------+---------+-------------+-----------------------+ | flags | object | REQUIRED | mb within mandate, ha | | | | | human approved, ev | | | | | action record written | +---------------+---------+-------------+-----------------------+ | mandate_hash | string | REQUIRED | SHA-256 of the | | | | | mandate in force | +---------------+---------+-------------+-----------------------+ | al | string | REQUIRED | Assurance level, AL0 | | | | | to AL3 | +---------------+---------+-------------+-----------------------+ | counter | integer | REQUIRED | Strictly increasing | | | | | per credential | +---------------+---------+-------------+-----------------------+ | action | object | REQUIRED | Action type and | | | | | target the signature | | | | | authorizes | Ganesan Expires 19 May 2027 [Page 14] Internet-Draft Agent Authenticator November 2026 +---------------+---------+-------------+-----------------------+ | evidence_head | string | OPTIONAL | Hash of the latest | | | | | action record | | | | | (Section 8) | +---------------+---------+-------------+-----------------------+ Table 7 6. Ceremonies This section defines five ceremonies. Of these, only registration and step-up require a person to take part. 6.1. Registration 1. The runtime asks the RP (or the IdP acting for it) to register an agent credential. The RP returns a nonce and its RP ID. 2. The AA creates a key pair scoped to that RP ID in secure hardware. 3. The AA builds the attestation statement (Section 5.4) bound to the nonce. 4. The AA builds a draft mandate and sends it to the human approver: locally as an OS prompt, remotely as a push or QR code to their phone. 5. The human reviews a plain-language rendering of the mandate and signs it with their passkey. 6. The RP verifies the attestation, the approver signature and the mandate, then stores the agent credential public key linked to the human's account, separately revocable. Ganesan Expires 19 May 2027 [Page 15] Internet-Draft Agent Authenticator November 2026 Human Agent Agent Relying approver Authenticator runtime party / IdP | | | | | | | 1. register | | | |--------------->| | | | 2. nonce, RP ID| | | |<---------------| | | 3. create key | | | |<---------------| | | |--+ 4. HW key, | | | |<-+ attestation | | | 5. draft mandate | | | |<-----------------| | | | 6. passkey signs | | | |----------------->| | | | | 7. key, attest.| | | | mandate | | | |--------------->| | | | | 8. submit | | | |--------------->| | | | |--+ 9. verify | | | |<-+ and store Figure 2: Registration 6.2. Authentication 1. The agent calls an RP through the runtime. The RP returns a challenge. 2. The runtime fills rp_id and origin from the connection and forwards the request to the AA. 3. The AA checks the mandate (RP allowed, action allowed, not expired, AL still met), writes an action record, and signs AgentData. 4. The RP verifies the assertion (Section 6.6), then issues a short- lived access token bound to the agent key with DPoP [RFC9449]. 6.3. Step-Up 1. The RP requires ha for an action in the mandate's step_up list, or the AA requires it by local policy. 2. The AA sends the action summary to the approver's authenticator (same device, or phone via push or hybrid transport). Ganesan Expires 19 May 2027 [Page 16] Internet-Draft Agent Authenticator November 2026 3. The human approves with a passkey assertion over SHA- 256(AgentData). 4. The AA attaches it and sets ha. The approval covers exactly that action and cannot be replayed. This ceremony is the same for local and cloud agents, because the approval is always made on the person's own authenticator. Agent Runtime Authenti- Human Relying | | cator (phone) party | | | | | | 1. call | | | | |-------->| 2. request | | | | |---------------------------------------->| | | 3. challenge | | |<----------------------------------------| | | 4. sign + | | | | | origin | | | | |------------>| | | | | |--+ 5. mandate | | | |<-+ check | | | | | 6. approve? | | step-up: | | |------------>| | risky | | | 7. passkey | | actions | | |<------------| | only | | 8. signed | | | | |<------------| | | | | 9. assertion | | |---------------------------------------->| | | 10. DPoP-bound token | | |<----------------------------------------| Figure 3: Authentication with step-up Steps 6 and 7 apply only to actions that require approval, such as merging, publishing or deleting. For other actions, the AA signs on the basis of the mandate alone. 6.4. Sub-Delegation An agent MAY hand a narrower task to another agent only if delegable is true. The parent AA signs a child mandate that MUST be a subset of the parent (RPs, actions, expiry) and MUST NOT lower min_al. Every hop's credential MUST be attested; the chain travels with the assertion and is verifiable offline. Ganesan Expires 19 May 2027 [Page 17] Internet-Draft Agent Authenticator November 2026 6.5. Revocation * Any of these MUST revoke an agent credential: mandate expiry, human or administrator revocation, IdP session revocation, failed re-attestation, broken action-record chain. * The AA is an OpenID Shared Signals Framework [SSF] receiver: a CAEP session-revoked or credential-change event from the IdP destroys the matching keys. * The AA is also a Shared Signals transmitter: it emits events when attestation fails or tampering is detected. * RPs MAY check credential status via a token status list for offline verification. 6.6. Relying Party Verification Procedure On receiving an assertion, the RP MUST perform these steps in order and MUST reject the request if any step fails: 1. Look up the agent credential by cred_id. Reject if unknown or revoked. 2. Verify that rp_id equals the RP's own RP ID and origin is an origin the RP serves. 3. Verify that challenge equals a nonce the RP issued, is unexpired, and has not been used before. 4. Verify the signature over AgentData with the stored credential public key. 5. Verify that counter is greater than the stored value, then store it. A non-increasing counter indicates a cloned credential; the RP MUST revoke it. 6. Verify that mandate_hash matches the mandate stored at registration and that the mandate has not expired. 7. Verify that action is permitted by the mandate's authorization_details. 8. Verify that al is at least the mandate's min_al and the RP's own policy minimum. Ganesan Expires 19 May 2027 [Page 18] Internet-Draft Agent Authenticator November 2026 9. If the action is in the mandate's step_up list or RP policy requires approval, verify that ha is set and that the attached WebAuthn assertion is a valid signature by the registered approver key over SHA-256(AgentData). 10. Only then issue a short-lived access token bound to the agent key with DPoP [RFC9449], or perform the action. An RP that delegates policy to an IdP (Section 7) MAY perform steps 6 to 9 by passing the assertion to the IdP in token exchange, but MUST still perform steps 1 to 5. 7. Protocol Bindings This section describes how the mechanism maps onto protocols that relying parties already support. In most cases the agent itself needs no changes, because the runtime and the AA handle authentication on its behalf. +============+=====================+================================+ | Binding | Where it plugs in | How the assertion is used | +============+=====================+================================+ | MCP | MCP authorization | AA acts as the MCP client's | | | [MCP-AUTH] and | auth provider; tokens are | | | its DPoP and | DPoP-bound to the agent key | | | workload identity | | | | extensions | | +------------+---------------------+--------------------------------+ | Enterprise | ID-JAG [ID-JAG]: | The user's identity assertion | | IdP | token exchange | is the subject token and the | | | [RFC8693] then | agent assertion is the actor | | | JWT bearer grant | token; the IdP applies its own | | | [RFC7523] | policy and issues the ID-JAG | +------------+---------------------+--------------------------------+ | Agent | Gateway | Gateway verifies the | | gateways | authorization on | assertion, reads al, ha and | | | each tool call | mandate hash, asks the IdP for | | | | the decision | +------------+---------------------+--------------------------------+ | Generic | JWT bearer or | RP swaps the assertion for a | | OAuth APIs | token exchange | short-lived, key-bound access | | | with | token | | | authorization | | | | details [RFC9396] | | +------------+---------------------+--------------------------------+ | Git over | git credential | Helper asks the AA for a token | | HTTPS | helper | per remote; no PATs on disk | +------------+---------------------+--------------------------------+ Ganesan Expires 19 May 2027 [Page 19] Internet-Draft Agent Authenticator November 2026 | Git over | ssh-agent | AA acts as an ssh-agent; | | SSH | protocol with a | signatures carry flags and | | | key type modeled | mandate hash | | | on OpenSSH FIDO | | | | keys | | +------------+---------------------+--------------------------------+ | Commit | SSH or Sigstore- | Commits signed by the agent | | signing | style signing | key; trailer names approver | | | | and mandate hash | +------------+---------------------+--------------------------------+ | Cloud | Workload identity | AA or IdP issues a token the | | | federation | cloud trusts; no static cloud | | | | keys | +------------+---------------------+--------------------------------+ | Package | Trusted | Publish requires ha; | | registries | publishing | provenance records agent | | | | identity | +------------+---------------------+--------------------------------+ | Agent | Pre-tool-use | Hook routes risky actions to | | hooks | hooks in coding | step-up before they run | | | agents | | +------------+---------------------+--------------------------------+ Table 8 Some relying parties support none of these mechanisms. For such an RP, the AA MAY fetch a legacy token through a broker, but MUST record al as no higher than AL1 and MUST NOT expose the token to the model process. 8. Action Records Action records are written and signed by the AA, using keys held in the same secure hardware as the agent's credentials. The agent cannot create or modify them. 8.1. What Is Recorded The AA records the following types of event: Action events: Every tool call or API call the AA signs or refuses, one record per action. Identity events: Registration, mandate signed, step-up approved or denied, revocation, re-attestation result. Local actions: File edits and shell commands reported by the runtime, labeled as runtime-reported. Ganesan Expires 19 May 2027 [Page 20] Internet-Draft Agent Authenticator November 2026 Integrity events: AA start and stop, heartbeats, chain verification failures. OS events (optional profile): Raw process, file and network activity, for deployments that need forensic depth. 8.2. Record Generation The AA writes one record per agent action at the moment it signs or refuses it. Because every request that needs a credential passes through the AA, the AA observes every action that the agent takes against a relying party. { "seq": 1042, "time": "2026-11-06T14:03:22Z", "agent": { "cred": "", "name": "coding-agent@laptop-17" }, "human": "priya@acme.example", "mandate_hash": "", "action": { "kind": "mcp.tool_call", "rp": "mcp.example", "tool": "jira.comment", "target": "PROJ-142", "args_hash": "sha256:" }, "decision": "allowed", "flags": { "mb": true, "ha": false }, "al": "AL2", "source": "daemon", "parent_seq": 1039, "prev_hash": "", "sig": "" } Ganesan Expires 19 May 2027 [Page 21] Internet-Draft Agent Authenticator November 2026 +======================+==============================+=========+ | Action | How the AA sees it | source | +======================+==============================+=========+ | MCP tool call to a | AA is the MCP client's auth | daemon | | remote server | provider and signs each call | | +----------------------+------------------------------+---------+ | API call (code host, | AA signs the request or the | daemon | | registry, cloud, | token exchange | | | model API) | | | +----------------------+------------------------------+---------+ | git push, commit | AA acts as git credential | daemon | | signing | helper and ssh-agent | | +----------------------+------------------------------+---------+ | Step-up request, | AA runs the ceremony | daemon | | approval or denial | | | +----------------------+------------------------------+---------+ | Action refused as | AA refuses to sign and | daemon | | outside the mandate | records why | | +----------------------+------------------------------+---------+ | Local tool (file | Runtime reports it to the AA | runtime | | edit, shell command, | through its hook | | | local MCP server) | | | +----------------------+------------------------------+---------+ Table 9 The decision member takes one of the following values: allowed, step_up_approved, step_up_denied, denied_out_of_scope, denied_origin_mismatch, denied_revoked. The parent_seq member ties each action to the one that led to it, so one task, including sub-agent hand-offs, can be reconstructed as a single sequence of actions, from the request that started the task to its last tool call. * Records are signed only by the AA. * Arguments are stored as a hash by default; full arguments stay in the local log. * Runtime-reported records MUST carry source: runtime. * The AA MUST emit a signed heartbeat at a fixed interval, so a stopped AA leaves a provable gap. 8.3. Tamper Evidence Ganesan Expires 19 May 2027 [Page 22] Internet-Draft Agent Authenticator November 2026 1. Each record carries the hash of the previous record, so edits, deletions and reordering are detectable. 2. The evidence key is held in secure hardware and rotates per epoch (forward-secure), so a later compromise cannot forge past records. 3. The evidence_head member of AgentData commits the chain to each RP, so RPs hold independent witnesses. 4. At each epoch, and on every ha event, the AA registers a signed statement with the chain head on a SCITT transparency service [RFC9943]. 8.4. SIEM Delivery Records are emitted as OCSF [OCSF] events so common SIEMs ingest them without a custom parser. Tool and API calls map to API Activity (6003), authentication and step-up to Authentication (3002), and mandate grants and revocations to Account Change (3001). 9. Deployment Profiles The protocol does not change between the two profiles below. What differs is how the AA is implemented and which assurance level it can reach. Ganesan Expires 19 May 2027 [Page 23] Internet-Draft Agent Authenticator November 2026 +============+=============================+========================+ | Aspect | Local profile | Cloud profile | | | (developer machine) | (sandbox, VM, enclave) | +============+=============================+========================+ | AA form | OS daemon in a separate | Sidecar, enclave key | | | privilege domain from | service, or attested | | | the agent | cloud KMS | +------------+-----------------------------+------------------------+ | Key | TPM 2.0, Secure Enclave | Confidential-computing | | storage | | enclave, vTPM, or KMS | | | | with attestation-gated | | | | release | +------------+-----------------------------+------------------------+ | Typical AL | AL2 (AL1 without secure | AL3 in confidential | | | hardware) | computing, AL1 with | | | | instance identity only | +------------+-----------------------------+------------------------+ | Human | Same device or phone | Always remote: phone | | approval | | via push or hybrid | | | | flow | +------------+-----------------------------+------------------------+ | Credential | Until mandate expiry or | One task; keys | | lifetime | revocation | destroyed when the | | | | environment ends | +------------+-----------------------------+------------------------+ | Key | AA MUST run outside the | AA MUST release keys | | isolation | agent's user or sandbox | only after attestation | | rule | and sign only for | matches the expected | | | processes it launched | measurement | +------------+-----------------------------+------------------------+ | Hand-off | A session moving to the | Same rule in reverse | | | cloud gets a new | | | | credential under a | | | | child mandate; it never | | | | copies the local key | | +------------+-----------------------------+------------------------+ Table 10 10. Implementation Status This section is to be removed before publishing as an RFC. This section records the status of known implementations of the protocol defined by this specification at the time of posting of this Internet-Draft, and is based on a proposal described in [RFC7942]. The description of implementations in this section is intended to assist the IETF in its decision processes in progressing drafts to Ganesan Expires 19 May 2027 [Page 24] Internet-Draft Agent Authenticator November 2026 RFCs. Please note that the listing of any individual implementation here does not imply endorsement by the IETF. Furthermore, no effort has been spent to verify the information presented here that was supplied by IETF contributors. This is not intended as, and must not be construed to be, a catalog of available implementations or their features. Readers are advised to note that other implementations may exist. According to [RFC7942], "this will allow reviewers and working groups to assign due consideration to documents that have the benefit of running code, which may serve as evidence of valuable experimentation and feedback that have made the implemented protocols more mature. It is up to the individual working groups to use this information as they see fit". At the time of writing, there are no known implementations. A reference implementation of the Agent Authenticator for Linux with TPM 2.0, for macOS with the Secure Enclave, and for one confidential- computing environment is planned. 11. Security Considerations The mechanism limits what a compromised agent can do and makes tampering with records detectable. It does not protect against a fully compromised host. Ganesan Expires 19 May 2027 [Page 25] Internet-Draft Agent Authenticator November 2026 +==================+=========================+=====================+ | Threat | Mitigation | Residual risk | +==================+=========================+=====================+ | Prompt injection | P3 origin binding; no | Injection can still | | sends agent to | credential exists for | misuse allowed RPs | | attacker origin | that origin | within the mandate | +------------------+-------------------------+---------------------+ | Injection abuses | Narrow mandates, step- | Damage up to | | an allowed RP | up for destructive | mandate scope | | | actions, action records | | +------------------+-------------------------+---------------------+ | Token theft from | Model never holds keys; | A token used from | | agent memory or | tokens DPoP-bound and | the same host | | logs | short-lived | before expiry | +------------------+-------------------------+---------------------+ | Same-user | AA in separate | Local administrator | | malware borrows | privilege domain, | can subvert the AA | | the AA | process-lineage check, | | | | ha for risky actions | | +------------------+-------------------------+---------------------+ | Administrator | Hash chain, forward- | Later events may be | | deletes or edits | secure hardware | missing, but the | | records | signing, RP-held heads, | gap is provable | | | SCITT receipts | | +------------------+-------------------------+---------------------+ | Approval fatigue | Plain-language | Humans can still | | | rendering of exactly | approve bad | | | what is signed; rate | requests | | | limits; risk-tiered ha | | +------------------+-------------------------+---------------------+ | Replay of | Approval signs SHA- | None known | | approvals | 256(AgentData) | | | | including RP challenge | | | | and counter | | +------------------+-------------------------+---------------------+ | Stale | Fresh nonce at | Window between re- | | attestation | registration; periodic | attestations | | | and per-ha re- | | | | attestation | | +------------------+-------------------------+---------------------+ | Over-broad sub- | Child mandate must be a | Depends on correct | | delegation | subset; min_al cannot | policy authoring | | | drop | | +------------------+-------------------------+---------------------+ Table 11 Ganesan Expires 19 May 2027 [Page 26] Internet-Draft Agent Authenticator November 2026 Three limitations should be noted. The record mechanisms detect tampering but do not prevent it. Guarantees on personal computers are weaker than those available in attested enclaves. Finally, assurance ultimately depends on the attestation roots of hardware and cloud vendors. 12. Privacy Considerations * Hardware attestation is linkable across RPs. RPs SHOULD receive only the normalized al, not raw vendor evidence, and AAs use per- RP credential keys (P3). * Action records contain sensitive data (file paths, commands, tool arguments). The local log stores full records; the SIEM receives configurable redaction; the transparency service receives only hashes. * Unlinkable presentation of al, for example using anonymous credentials such as BBS signatures or TPM Direct Anonymous Attestation, is left for future work. 13. IANA Considerations 13.1. Media Types IANA is requested to register the following media types [RFC6838] in the "Media Types" registry [IANA.MEDIA]. 13.1.1. application/agent-mandate+jwt Type name: application Subtype name: agent-mandate+jwt Required parameters: N/A Optional parameters: N/A Encoding considerations: binary; a JWT is a sequence of base64url- encoded values separated by period characters Security considerations: See Section 11 of this document Interoperability considerations: N/A Published specification: Section 5.1 of this document Applications that use this media type: Agent Authenticators, relying parties and identity providers that process agent mandates Fragment identifier considerations: N/A Additional information: Magic number(s): N/A; File extension(s): N/ A; Macintosh file type code(s): N/A Person and email address to contact for further information: Harish Santhanalakshmi Ganesan, harishsg99@gmail.com Intended usage: COMMON Restrictions on usage: none Ganesan Expires 19 May 2027 [Page 27] Internet-Draft Agent Authenticator November 2026 Author: Harish Santhanalakshmi Ganesan Change controller: IETF 13.1.2. application/agent-assertion+jwt Type name: application Subtype name: agent-assertion+jwt Required parameters: N/A Optional parameters: N/A Encoding considerations: binary; a JWT is a sequence of base64url- encoded values separated by period characters Security considerations: See Section 11 of this document Interoperability considerations: N/A Published specification: Section 5.3 of this document Applications that use this media type: Agent Authenticators, relying parties and identity providers that process agent assertions Fragment identifier considerations: N/A Additional information: Magic number(s): N/A; File extension(s): N/ A; Macintosh file type code(s): N/A Person and email address to contact for further information: Harish Santhanalakshmi Ganesan, harishsg99@gmail.com Intended usage: COMMON Restrictions on usage: none Author: Harish Santhanalakshmi Ganesan Change controller: IETF 13.2. OAuth URI IANA is requested to register the following value in the "OAuth URI" registry [IANA.OAUTH]. The agent assertion is used as the actor_token in token exchange [RFC8693], so that the agent is identified as acting on behalf of the user named by the subject token. URN: urn:ietf:params:oauth:token-type:agent-assertion Common Name: Agent assertion Change Controller: IETF Specification Document: Section 5.3 of this document 13.3. JSON Web Token Claims IANA is requested to register the following claims in the "JSON Web Token Claims" registry [IANA.JWT]. For each, the Change Controller is IETF and the Specification Document is Section 5.5 of this document. Ganesan Expires 19 May 2027 [Page 28] Internet-Draft Agent Authenticator November 2026 +===============+================================================+ | Claim Name | Claim Description | +===============+================================================+ | agent_cred | Agent credential, identified by key thumbprint | +---------------+------------------------------------------------+ | mandate_hash | Hash of the mandate in force | +---------------+------------------------------------------------+ | al | Agent environment assurance level | +---------------+------------------------------------------------+ | evidence_head | Hash of the latest action record | +---------------+------------------------------------------------+ Table 12 13.4. Agent Authenticator Registry Group IANA is requested to create a new registry group named "Agent Authenticator" containing the two registries below. Both use the "Specification Required" registration policy [RFC8126]. Designated experts should check that a proposed value is clearly defined, does not duplicate an existing value, and, for assurance levels, states the evidence a verifier needs to assign it. 13.4.1. Agent Assurance Levels Registry Template: Level name; Description; Reference. Initial contents: +=======+============================================+===========+ | Level | Description | Reference | +=======+============================================+===========+ | AL0 | Software key | This | | | | document | +-------+--------------------------------------------+-----------+ | AL1 | Platform keystore without hardware | This | | | isolation, or cloud workload identity only | document | +-------+--------------------------------------------+-----------+ | AL2 | Dedicated secure hardware; host software | This | | | not isolated | document | +-------+--------------------------------------------+-----------+ | AL3 | TEE that also isolates the agent's code, | This | | | with measured runtime | document | +-------+--------------------------------------------+-----------+ Table 13 Ganesan Expires 19 May 2027 [Page 29] Internet-Draft Agent Authenticator November 2026 13.4.2. Agent Action Decision Values Registry Template: Value; Description; Reference. Initial contents: +========================+======================+===========+ | Value | Description | Reference | +========================+======================+===========+ | allowed | Action signed within | This | | | the mandate | document | +------------------------+----------------------+-----------+ | step_up_approved | Human approved the | This | | | action | document | +------------------------+----------------------+-----------+ | step_up_denied | Human declined the | This | | | action | document | +------------------------+----------------------+-----------+ | denied_out_of_scope | Action outside the | This | | | mandate | document | +------------------------+----------------------+-----------+ | denied_origin_mismatch | Origin did not match | This | | | the credential scope | document | +------------------------+----------------------+-----------+ | denied_revoked | Credential or | This | | | mandate revoked | document | +------------------------+----------------------+-----------+ Table 14 14. References 14.1. Normative References [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, . [RFC7515] Jones, M., Bradley, J., and N. Sakimura, "JSON Web Signature (JWS)", RFC 7515, DOI 10.17487/RFC7515, May 2015, . [RFC7523] Jones, M., Campbell, B., and C. Mortimore, "JSON Web Token (JWT) Profile for OAuth 2.0 Client Authentication and Authorization Grants", RFC 7523, DOI 10.17487/RFC7523, May 2015, . Ganesan Expires 19 May 2027 [Page 30] Internet-Draft Agent Authenticator November 2026 [RFC7638] Jones, M. and N. Sakimura, "JSON Web Key (JWK) Thumbprint", RFC 7638, DOI 10.17487/RFC7638, September 2015, . [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017, . [RFC8693] Jones, M., Nadalin, A., Campbell, B., Ed., Bradley, J., and C. Mortimore, "OAuth 2.0 Token Exchange", RFC 8693, DOI 10.17487/RFC8693, January 2020, . [RFC9052] Schaad, J., "CBOR Object Signing and Encryption (COSE): Structures and Process", STD 96, RFC 9052, DOI 10.17487/RFC9052, August 2022, . [RFC9334] Birkholz, H., Thaler, D., Richardson, M., Smith, N., and W. Pan, "Remote ATtestation procedureS (RATS) Architecture", RFC 9334, DOI 10.17487/RFC9334, January 2023, . [RFC9396] Lodderstedt, T., Richer, J., and B. Campbell, "OAuth 2.0 Rich Authorization Requests", RFC 9396, DOI 10.17487/RFC9396, May 2023, . [RFC9449] Fett, D., Campbell, B., Bradley, J., Lodderstedt, T., Jones, M., and D. Waite, "OAuth 2.0 Demonstrating Proof of Possession (DPoP)", RFC 9449, DOI 10.17487/RFC9449, September 2023, . [RFC9711] Lundblade, L., Mandyam, G., O'Donoghue, J., and C. Wallace, "The Entity Attestation Token (EAT)", RFC 9711, DOI 10.17487/RFC9711, April 2025, . [WebAuthn] Cappalli, T., Ed., Kumar, A., Ed., Lundberg, E., Ed., Miller, M., Ed., Pascoe, Ed., and N. Satragno, Ed., "Web Authentication: An API for accessing Public Key Credentials Level 3", W3C REC-webauthn-3-20260825, 25 August 2026, . Latest version available at https://www.w3.org/TR/webauthn-3/. 14.2. Informative References Ganesan Expires 19 May 2027 [Page 31] Internet-Draft Agent Authenticator November 2026 [AIMS] Kasselman, P., Lombardo, J., Rosomakho, Y., Campbell, B., Steele, N., and A. Parecki, "AI Identity Management System", Work in Progress, Internet-Draft, draft-ietf- wimse-aims-00, September 2026, . [ASTRIX] Astrix Security, "Astrix Researchers Uncover Credential Risk in the Majority of MCP Servers", 2025, . [CSA-CI] Cloud Security Alliance, "Research Note: AI Coding Agent CI Prompt Injection", 7 August 2026, . [GG-SPRAWL] GitGuardian, "The State of Secrets Sprawl 2026", 2026, . [IANA.JWT] IANA, "JSON Web Token (JWT)", . [IANA.MEDIA] IANA, "Media Types", . [IANA.OAUTH] IANA, "OAuth Parameters", . [ID-JAG] Parecki, A., McGuinness, K., and B. Campbell, "Identity Assertion JWT Authorization Grant", Work in Progress, Internet-Draft, draft-ietf-oauth-identity-assertion-authz- grant-04, May 2026, . [INVARIANT] Invariant Labs, "GitHub MCP Exploited: Accessing private repositories via MCP", May 2025, . [MCP-AUTH] Model Context Protocol, "Model Context Protocol: Authorization", 2026, . Ganesan Expires 19 May 2027 [Page 32] Internet-Draft Agent Authenticator November 2026 [NX-LEAK] GitGuardian, "The Nx s1ngularity Attack: Inside the Credential Leak", 2025, . [OCSF] OCSF Project, "Open Cybersecurity Schema Framework", 2026, . [RFC6838] Freed, N., Klensin, J., and T. Hansen, "Media Type Specifications and Registration Procedures", BCP 13, RFC 6838, DOI 10.17487/RFC6838, January 2013, . [RFC7942] Sheffer, Y. and A. Farrel, "Improving Awareness of Running Code: The Implementation Status Section", BCP 205, RFC 7942, DOI 10.17487/RFC7942, July 2016, . [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for Writing an IANA Considerations Section in RFCs", BCP 26, RFC 8126, DOI 10.17487/RFC8126, June 2017, . [RFC9943] Birkholz, H., Delignat-Lavaud, A., Fournet, C., Deshpande, Y., and S. Lasker, "An Architecture for Trustworthy and Transparent Digital Supply Chains", RFC 9943, DOI 10.17487/RFC9943, April 2026, . [SHAI-HULUD] Elastic Security Labs, "Shai-Hulud Strikes Again: CHAINDROP Worm Hits 400+ npm Packages", 2026, . [SSF] Tulshibagwale, A., Cappalli, T., Scurtescu, M., Backman, A., Bradley, J., and S. Miel, "OpenID Shared Signals Framework Specification 1.0", 29 August 2025, . [VB-EXPLOITS] VentureBeat, "Six exploits broke AI coding agents. IAM never saw them", 2026, . Ganesan Expires 19 May 2027 [Page 33] Internet-Draft Agent Authenticator November 2026 Appendix A. Standardization Path This section is to be removed before publishing as an RFC. The specification may be split into smaller documents, each sent to the group that owns that layer. +=====================+================+===========================+ | Document | Venue | Contents | +=====================+================+===========================+ | EAT profile for | IETF RATS | Vendor evidence to | | agent keys | | AL0-AL3 mapping | +---------------------+----------------+---------------------------+ | Agent assertion and | IETF OAuth, | AgentData, mandate, token | | mandate profile | with WIMSE | exchange and ID-JAG use | +---------------------+----------------+---------------------------+ | Delegated approval | W3C WebAuthn | Passkey signing of | | extension | WG, FIDO | mandates and per-action | | | Alliance | approvals | +---------------------+----------------+---------------------------+ | Agent action-record | IETF SCITT, | Records, chain, anchoring | | profile | plus OCSF | and OCSF mapping | +---------------------+----------------+---------------------------+ | MCP binding | MCP | DPoP and workload | | | authorization | identity usage with the | | | working groups | AA | +---------------------+----------------+---------------------------+ Table 15 Appendix B. Related Work The AI Identity Management System [AIMS] composes workload identity, OAuth and shared signals into a model for agent authentication and authorization, treating agents as workloads; this document adds hardware-bound agent keys, runtime-enforced origin binding and human approval with passkeys. The Identity Assertion JWT Authorization Grant [ID-JAG] and MCP authorization [MCP-AUTH] let identity providers control which applications an agent may reach; this document supplies them with stronger evidence about the agent rather than replacing them. Transparency services [RFC9943] provide the anchoring used for action records. Appendix C. Acknowledgments This draft builds on the WebAuthn, OAuth, RATS, WIMSE and SCITT work of the W3C, the FIDO Alliance and the IETF. Ganesan Expires 19 May 2027 [Page 34] Internet-Draft Agent Authenticator November 2026 Author's Address Harish Santhanalakshmi Ganesan Email: harishsg99@gmail.com Ganesan Expires 19 May 2027 [Page 35]