Files

Christian Nennemann d8d1524dac Make SPIFFE ID format recommended, not required for iss claim

Allow any URI scheme for the iss claim (SPIFFE, HTTPS, URN:UUID)
to support non-WIMSE deployments that want DAG tracing without
SPIFFE infrastructure. SPIFFE format remains SHOULD for WIMSE
deployments.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-02-24 23:27:34 +01:00

72 KiB

Raw Blame History

fullname: Christian Nennemann
organization: Independent Researcher
email: ietf@nennemann.de

normative: RFC7515: RFC7517: RFC7519: RFC7518: RFC9562: RFC9110: I-D.ietf-wimse-arch: I-D.ietf-wimse-s2s-protocol:

informative: RFC3552: RFC8693: RFC9421: I-D.ni-wimse-ai-agent-identity: SPIFFE: title: "Secure Production Identity Framework for Everyone (SPIFFE)" target: https://spiffe.io/docs/latest/spiffe-about/overview/ date: false EU-AI-ACT: title: "Regulation (EU) 2024/1689 of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)" target: https://eur-lex.europa.eu/eli/reg/2024/1689 date: 2024-06-13 author: - org: European Parliament and Council of the European Union FDA-21CFR11: title: "Title 21, Code of Federal Regulations, Part 11: Electronic Records; Electronic Signatures" target: https://www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-11 date: false author: - org: U.S. Food and Drug Administration MIFID-II: title: "Directive 2014/65/EU of the European Parliament and of the Council on markets in financial instruments (MiFID II)" target: https://eur-lex.europa.eu/eli/dir/2014/65 date: 2014-05-15 author: - org: European Parliament and Council of the European Union DORA: title: "Regulation (EU) 2022/2554 on digital operational resilience for the financial sector (DORA)" target: https://eur-lex.europa.eu/eli/reg/2022/2554 date: 2022-12-14 author: - org: European Parliament and Council of the European Union EU-MDR: title: "Regulation (EU) 2017/745 on medical devices (MDR)" target: https://eur-lex.europa.eu/eli/reg/2017/745 date: 2017-04-05 author: - org: European Parliament and Council of the European Union OPENTELEMETRY: title: "OpenTelemetry Specification" target: https://opentelemetry.io/docs/specs/otel/ date: false author: - org: Cloud Native Computing Foundation I-D.ietf-scitt-architecture: I-D.ietf-oauth-transaction-tokens: I-D.oauth-transaction-tokens-for-agents:

--- abstract

This document defines Execution Context Tokens (ECTs), an extension to the Workload Identity in Multi System Environments (WIMSE) architecture for distributed agentic workflows in regulated environments. ECTs provide signed, structured records of task execution order, policy evaluation outcomes, and compliance state across agent-to-agent communication. By extending WIMSE Workload Identity Tokens with execution context claims in JSON Web Token (JWT) format, this specification enables regulated systems to maintain structured audit trails that support compliance verification. ECTs use a directed acyclic graph (DAG) structure to represent task dependencies, record policy evaluation outcomes at each decision point, and integrate with WIMSE Workload Identity Tokens (WIT) using the same signing model and cryptographic primitives. A new HTTP header field, Execution-Context, is defined for transporting ECTs alongside existing WIMSE headers. ECTs are a technical building block that supports, but does not by itself constitute, compliance with regulatory frameworks.

--- middle

Introduction

Motivation

The Workload Identity in Multi System Environments (WIMSE) framework {{I-D.ietf-wimse-arch}} provides robust workload authentication through Workload Identity Tokens (WIT) and Workload Proof Tokens (WPT). The WIMSE service-to-service protocol {{I-D.ietf-wimse-s2s-protocol}} defines how workloads authenticate each other across call chains using the Workload-Identity and Workload-Proof-Token HTTP headers.

However, workload identity alone does not address execution accountability. Knowing who performed an action does not record what was done, what policy was applied, or whether compliance requirements were evaluated at each decision point.

Regulated environments increasingly deploy autonomous agents that coordinate across organizational boundaries. Multiple regulatory frameworks — including {{EU-AI-ACT}}, {{FDA-21CFR11}}, {{MIFID-II}}, and {{DORA}} — require structured, auditable records of automated decision-making and execution (see {{table-regulatory}} for a detailed mapping).

This document defines an extension to the WIMSE architecture that addresses the gap between workload identity and execution accountability. WIMSE authenticates agents; this extension records what they did, in what order, and what policy was evaluated.

As identified in {{I-D.ni-wimse-ai-agent-identity}}, call context in agentic workflows needs to be visible and preserved. ECTs provide a mechanism to address this requirement with cryptographic assurances.

Problem Statement

Three core gaps exist in current approaches to regulated agentic systems:

WIMSE authenticates agents but does not record what they actually did. A WIT proves "Agent A is authorized" but not "Agent A executed Task X, under Policy Y, producing Output Z."
No standard mechanism exists to record policy evaluation outcomes at each decision point in a multi-agent workflow.
No mechanism exists to cryptographically link compensation and rollback decisions to original actions.

Existing observability tools such as distributed tracing {{OPENTELEMETRY}} provide visibility for debugging and monitoring but do not provide cryptographic assurances. Tracing data is not cryptographically signed, not tamper-evident, and not designed for regulatory audit scenarios.

Scope and Applicability

This document defines:

The Execution Context Token (ECT) format ({{ect-format}})
DAG structure for task dependency ordering ({{dag-validation}})
Policy checkpoint recording ({{policy-claims}})
Integration with the WIMSE identity framework ({{wimse-integration}})
An HTTP header for ECT transport ({{http-header}})
Audit ledger interface requirements ({{ledger-interface}})

The following are out of scope and are handled by WIMSE:

Workload authentication and identity provisioning
Key distribution and management
Trust domain establishment and management
Credential lifecycle management

Relationship to Regulatory Compliance

ECTs are a technical mechanism that can support compliance programs by providing structured, cryptographically signed execution records. ECTs do not by themselves constitute compliance with any regulatory framework referenced in this document.

Compliance with each referenced regulation requires organizational controls, policies, procedures, validation, and governance measures beyond the scope of this specification. The regulatory references in this document are intended to motivate the design requirements, not to claim that implementing ECTs satisfies these regulations.

ECTs provide evidence of claimed execution ordering and policy evaluation. They do not independently verify that the claimed execution actually occurred as described, that the policy evaluation was correct, or that the agent faithfully performed the stated action. The trustworthiness of ECT claims depends on the trustworthiness of the signing agent and the integrity of the broader deployment environment.

Conventions and Definitions

{::boilerplate bcp14-tagged}

The following terms are used in this document:

Agent:: An autonomous workload, as defined by WIMSE {{I-D.ietf-wimse-arch}}, that executes tasks within a workflow.
Task:: A discrete unit of agent work that consumes inputs and produces outputs.
Directed Acyclic Graph (DAG):: A graph structure representing task dependency ordering where edges are directed and no cycles exist.
Execution Context Token (ECT):: A JSON Web Token {{RFC7519}} defined by this specification that records task execution details and policy evaluation outcomes.
Audit Ledger:: An append-only, immutable log of all ECTs within a workflow or set of workflows, used for regulatory audit and compliance verification.
Policy Checkpoint:: A point in a workflow where a policy evaluation outcome is recorded within an ECT.
Workload Identity Token (WIT):: A WIMSE credential proving a workload's identity within a trust domain.
Workload Proof Token (WPT):: A WIMSE proof-of-possession token used for request-level authentication.
Trust Domain:: A WIMSE concept representing an organizational boundary with a shared identity issuer, corresponding to a SPIFFE {{SPIFFE}} trust domain.
Witness:: A third-party entity that observes and attests to the execution of a task, providing additional accountability.

WIMSE Architecture Integration

WIMSE Foundation

The WIMSE architecture {{I-D.ietf-wimse-arch}} defines:

Workload Identity Tokens (WIT) that prove a workload's identity within a trust domain ("I am Agent X in trust domain Y")
Workload Proof Tokens (WPT) that prove possession of the private key associated with a WIT ("I control the key for Agent X")
Multi-hop authentication via the service-to-service protocol {{I-D.ietf-wimse-s2s-protocol}}

The following execution accountability needs are complementary to the WIMSE scope and are not addressed by workload identity alone:

Recording what agents actually do with their authenticated identity
Recording policy evaluation outcomes at each hop
Maintaining structured execution records
Linking compensation or rollback actions to original tasks

Extension Model

ECTs extend WIMSE by adding an execution accountability layer between the identity layer and the application layer:

+--------------------------------------------------+
|  WIMSE Layer (Identity)                          |
|    WIT: "I am Agent X (spiffe://td/agent/x)"    |
|    WPT: "I prove I control the key for Agent X"  |
+--------------------------------------------------+
                       |
                       v
+--------------------------------------------------+
|  ECT Layer (Execution Accountability) [This Spec]|
|    ECT: "Task executed, dependencies met,        |
|          policy evaluated, outcome recorded"      |
+--------------------------------------------------+
                       |
                       v
+--------------------------------------------------+
|  Ledger Layer (Immutable Record)                 |
|    "All ECTs appended to audit ledger"           |
+--------------------------------------------------+

{: #fig-layers title="WIMSE Extension Architecture Layers"}

This extension reuses the WIMSE signing model, extends JWT claims using standard JWT extensibility {{RFC7519}}, and maintains WIMSE concepts including trust domains and workload identifiers.

Integration Points

An ECT integrates with the WIMSE identity framework through the following mechanisms:

The ECT JOSE header "kid" parameter MUST reference the public key identifier from the agent's WIT.
In WIMSE deployments, the ECT "iss" claim SHOULD use the WIMSE workload identifier format (a SPIFFE ID {{SPIFFE}}).
The ECT MUST be signed with the same private key associated with the agent's WIT.
The ECT signing algorithm (JOSE header "alg" parameter) MUST match the algorithm used in the corresponding WIT.

When an agent makes an HTTP request to another agent, the Execution-Context header is carried alongside WIMSE identity headers:

HTTP Request from Agent A to Agent B:
  Workload-Identity:    <WIT for Agent A>
  Execution-Context:    <ECT recording what A did>

{: #fig-http-headers title="HTTP Header Stacking"}

When a Workload Proof Token (WPT) is available per {{I-D.ietf-wimse-s2s-protocol}}, agents SHOULD include it alongside the WIT and ECT. ECT verification does not depend on the presence of a WPT; the ECT is independently verifiable via the WIT public key.

The receiving agent (Agent B) verifies in order:

WIT (WIMSE layer): Verifies Agent A's identity within the trust domain. WPT verification, if present, per {{I-D.ietf-wimse-s2s-protocol}}.
ECT (this extension): Records what Agent A did, what policy was evaluated, and what precedent tasks exist.
Ledger: Appends the verified ECT to the audit ledger.

Execution Context Token Format

An Execution Context Token is a JSON Web Token (JWT) {{RFC7519}} signed as a JSON Web Signature (JWS) {{RFC7515}} using the Compact Serialization. JWS JSON Serialization MUST NOT be used for ECTs.

JOSE Header

The ECT JOSE header MUST contain the following parameters:

{
  "alg": "ES256",
  "typ": "wimse-exec+jwt",
  "kid": "agent-a-key-id-123"
}

{: #fig-header title="ECT JOSE Header Example"}

alg:: REQUIRED. The digital signature algorithm used to sign the ECT. MUST match the algorithm in the corresponding WIT. Implementations MUST support ES256 {{RFC7518}}. The "alg" value MUST NOT be "none". Symmetric algorithms (e.g., HS256, HS384, HS512) MUST NOT be used, as ECTs require asymmetric signatures for non-repudiation.
typ:: REQUIRED. MUST be set to "wimse-exec+jwt" to distinguish ECTs from other JWT types, consistent with the WIMSE convention for type parameter values.
kid:: REQUIRED. The key identifier referencing the public key from the agent's WIT {{RFC7517}}. Used by verifiers to look up the correct public key for signature verification.

JWT Claims

The ECT payload contains both WIMSE-compatible standard JWT claims and execution context claims defined by this specification.

Standard JWT Claims

The following standard JWT claims {{RFC7519}} MUST be present in every ECT:

iss:

REQUIRED. StringOrURI. A URI identifying the issuer of the ECT. In WIMSE deployments, this SHOULD be the workload's SPIFFE ID in the format spiffe://<trust-domain>/<path>, matching the "sub" claim of the agent's WIT. Non-WIMSE deployments MAY use other URI schemes (e.g., HTTPS URLs or URN:UUID identifiers).

sub:

OPTIONAL. StringOrURI. The subject of the ECT. When present, MUST equal the "iss" claim. This claim is included for compatibility with JWT libraries and frameworks that expect a "sub" claim to be present.

aud:

REQUIRED. StringOrURI or array of StringOrURI. The intended recipient(s) of the ECT. Because ECTs serve as both inter-agent messages and audit records, the "aud" claim SHOULD contain the identifiers of all entities that will verify the ECT. In practice this means:

Point-to-point delivery: when an ECT is sent from one agent to a single next agent, "aud" contains that agent's workload identity. The receiving agent verifies the ECT and forwards it to the ledger on behalf of the issuer.
Direct-to-ledger submission: when an ECT is submitted directly to the audit ledger (e.g., after a join or at workflow completion), "aud" contains the ledger's identity.
Multi-audience: when an ECT must be verified by both the next agent and the ledger independently, "aud" MUST be an array containing both identifiers (e.g., ["spiffe://example.com/agent/next", "spiffe://example.com/system/ledger"]). Each verifier checks that its own identity appears in the array.

In multi-parent (join) scenarios where a task depends on ECTs from multiple parent agents, the joining agent creates a new ECT with the parent task IDs in "par". The "aud" of this new ECT is set according to the rules above based on where the ECT will be delivered — it is independent of the "aud" values in the parent ECTs.

iat:

REQUIRED. NumericDate. The time at which the ECT was issued. The ECT records a completed action, so the "iat" value reflects when the record was created, not when task execution began.

exp:

REQUIRED. NumericDate. The expiration time of the ECT. Implementations SHOULD set this to 5 to 15 minutes after "iat" to limit the replay window while allowing for reasonable clock skew and processing time.

The standard JWT "nbf" (Not Before) claim is not used in ECTs because ECTs record completed actions and are valid immediately upon issuance.

jti:: REQUIRED. String. A globally unique identifier for both the ECT and the task it records, in UUID format {{RFC9562}}. Since each ECT represents exactly one task, "jti" serves as both the token identifier (for replay detection) and the task identifier (for DAG parent references in "par"). Receivers MUST reject ECTs whose "jti" has already been seen within the expiration window. When "wid" is present, uniqueness is scoped to the workflow; when "wid" is absent, uniqueness MUST be enforced globally across the ECT store.

Execution Context

The following claims are defined by this specification:

wid:: OPTIONAL. String. A workflow identifier that groups related ECTs into a single workflow. When present, MUST be a UUID {{RFC9562}}.
exec_act:: REQUIRED. String. The action or task type identifier describing what the agent performed (e.g., "process_payment", "validate_safety", "calculate_dosage"). Note: this claim is intentionally named "exec_act" rather than "act" to avoid collision with the "act" (Actor) claim registered by {{RFC8693}}.
par:: REQUIRED. Array of strings. Parent task identifiers representing DAG dependencies. Each element MUST be the "jti" value of a previously verified ECT. An empty array indicates a root task with no dependencies. A workflow MAY contain multiple root tasks.

Policy Evaluation

The following claims record policy evaluation outcomes:

pol:

OPTIONAL. String. The identifier of the policy rule that was evaluated for this task (e.g., "clinical_data_access_policy_v1"). MUST be present when "pol_decision" is present.

pol_decision:

OPTIONAL. String. The result of the policy evaluation. When present, MUST be one of the values registered in the ECT Policy Decision Values registry ({{pol-decision-registry}}). MUST be present when "pol" is present. Initial values are:

"approved": The policy evaluation succeeded and the task was authorized to proceed.
"rejected": The policy evaluation failed. A "rejected" ECT MUST still be recorded for accountability. An ECT with "pol_decision" of "rejected" MAY appear as a parent in the "par" array of a subsequent ECT, but only for compensation, rollback, or remediation tasks. Agents MUST NOT proceed with normal workflow execution based on a parent ECT whose "pol_decision" is "rejected".
"pending_human_review": The policy evaluation requires human judgment before proceeding. Agents MUST NOT proceed with dependent tasks until a subsequent ECT from a human reviewer records an "approved" decision referencing this task as a parent.

When "pol" and "pol_decision" are absent, the ECT records task execution without a policy checkpoint. Regulated deployments SHOULD include policy claims on all ECTs to maintain complete audit trails.

pol_enforcer:

OPTIONAL. StringOrURI. The identity of the entity (system or person) that evaluated the policy decision. When present, SHOULD use SPIFFE ID format.

This specification intentionally defines only the recording of policy evaluation outcomes. The mechanisms by which policies are defined, distributed to agents, and evaluated are out of scope. The "pol" claim is an opaque identifier referencing an external policy; the semantics and enforcement of that policy are determined by the deployment environment. Implementations may use any policy engine or framework (e.g., OPA/Rego, Cedar, XACML, or custom solutions) provided that the evaluation outcome is faithfully recorded in the ECT claims defined above.

Data Integrity

The following claims provide integrity verification for task inputs and outputs without revealing the data itself:

inp_hash:: OPTIONAL. String. A cryptographic hash of the input data, formatted as "hash-algorithm:base64url-encoded-hash" (e.g., "sha-256:n4bQgYhMfWWaL-qgxVrQFaO_TxsrC4Is0V1sFbDwCgg"). The hash algorithm identifier MUST be a lowercase value from the IANA Named Information Hash Algorithm Registry (e.g., "sha-256", "sha-384", "sha-512"). Implementations MUST support "sha-256" and SHOULD use "sha-256" unless a stronger algorithm is required. Implementations MUST NOT accept hash algorithms weaker than SHA-256 (e.g., MD5, SHA-1). The hash MUST be computed over the raw octets of the input data.
out_hash:: OPTIONAL. String. A cryptographic hash of the output data, using the same format and algorithm requirements as "inp_hash".

Compensation and Rollback

compensation_required:: OPTIONAL. Boolean. Indicates whether this task is a compensation or rollback action for a previous task.
compensation_reason:: OPTIONAL. String. A human-readable reason for the compensation action. MUST be present if "compensation_required" is true. Values SHOULD use structured identifiers (e.g., "policy_violation_in_parent_trade") rather than free-form text to minimize the risk of embedding sensitive information. See {{data-minimization}} for privacy guidance. If "compensation_reason" is present, "compensation_required" MUST be true.

Note: compensation ECTs reference historical parent tasks via the "par" claim. The referenced parent ECTs may have passed their own "exp" time; ECT expiration applies to the verification window of the ECT itself, not to its validity as a parent reference in the ledger.

Extensions

ext:: OPTIONAL. Object. An extension object for domain-specific claims not defined by this specification. Implementations that do not understand extension claims MUST ignore them.

To avoid key collisions between different domains, extension key names MUST use reverse domain notation (e.g., "com.example.custom_field"). Implementations MUST NOT use unqualified key names within the "ext" object. To prevent abuse and excessive token size, the serialized JSON representation of the "ext" object SHOULD NOT exceed 4096 bytes, and the JSON nesting depth within the "ext" object SHOULD NOT exceed 5 levels. Implementations SHOULD reject ECTs whose "ext" claim exceeds these limits.

The following extension keys are RECOMMENDED for common use cases. These are not registered claims; they are carried within the "ext" object:

"org.ietf.wimse.exec_time_ms": Integer. Execution duration in milliseconds.
"org.ietf.wimse.regulated_domain": String. Regulatory domain (e.g., "medtech", "finance", "military").
"org.ietf.wimse.model_version": String. AI/ML model version.
"org.ietf.wimse.witnessed_by": Array of StringOrURI. Identifiers of third-party entities that the issuer claims observed the task. Note: this is self-asserted; for verifiable witness attestation, witnesses should submit independent signed ECTs.
"org.ietf.wimse.inp_classification": String. Data sensitivity classification (e.g., "public", "confidential", "restricted").
"org.ietf.wimse.pol_timestamp": NumericDate. Time at which the policy decision was made, if distinct from "iat".

Complete ECT Example

The following is a complete ECT payload example:

{
  "iss": "spiffe://example.com/agent/clinical",
  "sub": "spiffe://example.com/agent/clinical",
  "aud": "spiffe://example.com/agent/safety",
  "iat": 1772064150,
  "exp": 1772064750,
  "jti": "550e8400-e29b-41d4-a716-446655440001",

  "wid": "a0b1c2d3-e4f5-6789-abcd-ef0123456789",
  "exec_act": "recommend_treatment",
  "par": [],

  "pol": "clinical_reasoning_policy_v2",
  "pol_decision": "approved",
  "pol_enforcer": "spiffe://example.com/policy/clinical-engine",

  "inp_hash": "sha-256:n4bQgYhMfWWaL-qgxVrQFaO_TxsrC4Is0V1sFbDwCgg",
  "out_hash": "sha-256:LCa0a2j_xo_5m0U8HTBBNBNCLXBkg7-g-YpeiGJm564",

  "ext": {
    "org.ietf.wimse.pol_timestamp": 1772064145,
    "org.ietf.wimse.exec_time_ms": 245,
    "org.ietf.wimse.regulated_domain": "medtech",
    "org.ietf.wimse.model_version": "clinical-reasoning-v4.2"
  }
}

{: #fig-full-ect title="Complete ECT Payload Example"}

HTTP Header Transport

Execution-Context Header Field

This specification defines the Execution-Context HTTP header field {{RFC9110}} for transporting ECTs between agents.

The header field value is the ECT in JWS Compact Serialization format {{RFC7515}}. The value consists of three Base64url-encoded parts separated by period (".") characters.

An agent sending a request to another agent includes the Execution-Context header alongside the WIMSE Workload-Identity header:

GET /api/safety-check HTTP/1.1
Host: safety-agent.example.com
Workload-Identity: eyJhbGci...WIT...
Execution-Context: eyJhbGci...ECT...

{: #fig-http-example title="HTTP Request with ECT Header"}

When multiple parent tasks contribute context to a single request, multiple Execution-Context header field lines MAY be included, each carrying a separate ECT in JWS Compact Serialization format.

When a receiver processes multiple Execution-Context headers, it MUST individually verify each ECT per the procedure in {{verification}}. If any single ECT fails verification, the receiver MUST reject the entire request. The set of verified parent task IDs across all received ECTs represents the complete set of parent dependencies available for the receiving agent's subsequent ECT.

DAG Validation

Overview

ECTs form a Directed Acyclic Graph (DAG) where each task references its parent tasks via the "par" claim. This structure provides a cryptographically signed record of execution ordering, enabling auditors to reconstruct the complete workflow and verify that required predecessor tasks were recorded before dependent tasks.

DAG validation is performed against the audit ledger, which serves as the authoritative store of previously verified ECTs.

Validation Rules

When receiving and verifying an ECT, implementations MUST perform the following DAG validation steps:

Task ID Uniqueness: The "jti" claim MUST be unique within the applicable scope (the workflow identified by "wid", or the entire ECT store if "wid" is absent). If an ECT with the same "jti" already exists, the ECT MUST be rejected.
Parent Existence: Every task identifier listed in the "par" array MUST correspond to a task that is available in the ECT store (either previously recorded in the ledger or received inline as a verified parent ECT). If any parent task is not found, the ECT MUST be rejected.
Temporal Ordering: The "iat" value of every parent task MUST NOT be greater than the "iat" value of the current task plus a configurable clock skew tolerance (RECOMMENDED: 30 seconds). That is, for each parent: parent.iat < child.iat + clock_skew_tolerance. The tolerance accounts for clock skew between agents; it does not guarantee strict causal ordering from timestamps alone. Causal ordering is primarily enforced by the DAG structure (parent existence in the ECT store), not by timestamps. If any parent task violates this constraint, the ECT MUST be rejected.
Acyclicity: Following the chain of parent references MUST NOT lead back to the current ECT's "jti". If a cycle is detected, the ECT MUST be rejected.
Parent Policy Decision: If any parent ECT contains a "pol_decision" of "rejected" or "pending_human_review", the current ECT's "exec_act" MUST indicate a compensation, rollback, remediation, or human review action. Implementations MUST NOT accept an ECT representing normal workflow continuation when a parent's "pol_decision" is not "approved", unless the current ECT has "compensation_required" set to true. This rule only applies when the parent ECT contains policy claims.
Trust Domain Consistency: Parent tasks SHOULD belong to the same trust domain or to a trust domain with which a federation relationship has been established.

DAG Validation Algorithm

The following pseudocode describes the DAG validation procedure:

function validate_dag(ect, ledger, clock_skew_tolerance):
  // Step 1: Uniqueness check
  if ledger.contains(ect.jti, ect.wid):
    return error("ECT ID already exists")

  // Step 2: Parent existence and temporal ordering
  for parent_id in ect.par:
    parent = ledger.get(parent_id)
    if parent is null:
      return error("Parent task not found: " + parent_id)
    if parent.iat >= ect.iat + clock_skew_tolerance:
      return error("Parent task not earlier than current")

  // Step 3: Cycle detection (with traversal limit)
  visited = set()
  result = has_cycle(ect.jti, ect.par, ledger, visited,
                      max_ancestor_limit)
  if result is error or result is true:
    return error("Circular dependency or depth limit exceeded")

  return success

function has_cycle(target_jti, parent_ids, ledger,
                   visited, max_depth):
  if visited.size() >= max_depth:
    return error("Maximum ancestor traversal limit exceeded")
  for parent_id in parent_ids:
    if parent_id == target_jti:
      return true
    if parent_id in visited:
      continue
    visited.add(parent_id)
    parent = ledger.get(parent_id)
    if parent is not null:
      result = has_cycle(target_jti, parent.par, ledger,
                          visited, max_depth)
      if result is error or result is true:
        return result
  return false

{: #fig-dag-validation title="DAG Validation Pseudocode"}

The cycle detection traverses the ancestor graph rooted at the current task's parents. The complexity is O(V) where V is the number of ancestor nodes reachable from the current task's parent references. For typical workflows with shallow DAGs, this is efficient. To prevent denial-of-service via extremely deep or wide DAGs, implementations SHOULD enforce a maximum ancestor traversal limit (RECOMMENDED: 10000 nodes). If the limit is reached before cycle detection completes, the ECT SHOULD be rejected. Implementations SHOULD cache cycle detection results for previously verified tasks to avoid redundant traversals.

Signature and Token Verification

Verification Procedure

When an agent receives an ECT, it MUST perform the following verification steps in order:

Parse the JWS Compact Serialization to extract the JOSE header, payload, and signature components per {{RFC7515}}.
Verify that the "typ" header parameter is "wimse-exec+jwt".
Verify that the "alg" header parameter is not "none" and is not a symmetric algorithm.
Verify the "kid" header parameter references a known, valid public key from a WIT within the trust domain.
Retrieve the public key identified by "kid" and verify the JWS signature per {{RFC7515}} Section 5.2.
Verify that the signing key identified by "kid" has not been revoked within the trust domain. Implementations MUST check the key's revocation status using the trust domain's key lifecycle mechanism (e.g., certificate revocation list, OCSP, or SPIFFE trust bundle updates).
Verify the "alg" header parameter matches the algorithm in the corresponding WIT.
Verify the "iss" claim matches the "sub" claim of the WIT associated with the "kid" public key.
Verify the "aud" claim contains the verifier's own workload identity. When "aud" is an array, it is sufficient that the verifier's identity appears as one element; the presence of other audience values does not cause verification failure. When the verifier is the audit ledger, the ledger's own identity MUST appear in "aud".
Verify the "exp" claim indicates the ECT has not expired.
Verify the "iat" claim is not unreasonably far in the past (implementation-specific threshold, RECOMMENDED maximum of 15 minutes) and is not unreasonably far in the future (RECOMMENDED: no more than 30 seconds ahead of the verifier's current time, to account for clock skew).
Verify all required claims ("jti", "exec_act", "par") are present and well-formed.
If "pol" or "pol_decision" is present, verify that both are present and that "pol_decision" is one of "approved", "rejected", or "pending_human_review".
Perform DAG validation per {{dag-validation}}.
If all checks pass, the ECT MUST be appended to the audit ledger.

If any verification step fails, the ECT MUST be rejected and the failure MUST be logged for audit purposes. Error messages SHOULD NOT reveal whether specific parent task IDs exist in the ledger, to prevent information disclosure.

When ECT verification fails during HTTP request processing, the receiving agent SHOULD respond with HTTP 403 (Forbidden) if the WIT is valid but the ECT is invalid, and HTTP 401 (Unauthorized) if the ECT signature verification fails. The response body SHOULD include a generic error indicator without revealing which specific verification step failed. The receiving agent MUST NOT process the requested action when ECT verification fails.

Verification Pseudocode

function verify_ect(ect_jws, verifier_id,
                     trust_domain_keys, ledger):
  // Parse JWS
  (header, payload, signature) = parse_jws(ect_jws)

  // Verify header
  if header.typ != "wimse-exec+jwt":
    return reject("Invalid typ parameter")

  if header.alg == "none" or is_symmetric(header.alg):
    return reject("Prohibited algorithm")

  // Look up public key
  public_key = trust_domain_keys.get(header.kid)
  if public_key is null:
    return reject("Unknown key identifier")

  // Verify signature
  if not verify_jws_signature(header, payload,
                               signature, public_key):
    return reject("Invalid signature")

  // Verify key not revoked
  if is_key_revoked(header.kid, trust_domain_keys):
    return reject("Signing key has been revoked")

  // Verify algorithm alignment
  wit = get_wit_for_key(header.kid)
  if header.alg != wit.alg:
    return reject("Algorithm mismatch with WIT")

  // Verify issuer matches WIT subject
  if payload.iss != wit.sub:
    return reject("Issuer does not match WIT subject")

  // Verify audience
  if verifier_id not in payload.aud:
    return reject("ECT not intended for this recipient")

  // Verify not expired
  if payload.exp < current_time():
    return reject("ECT has expired")

  // Verify iat freshness (not too old, not in the future)
  if payload.iat < current_time() - max_age_threshold:
    return reject("ECT issued too long ago")
  if payload.iat > current_time() + clock_skew_tolerance:
    return reject("ECT issued in the future")

  // Verify required claims
  for claim in ["jti", "exec_act", "par"]:
    if claim not in payload:
      return reject("Missing required claim: " + claim)

  // Validate policy claims (optional, but must be paired)
  if "pol" in payload or "pol_decision" in payload:
    if "pol" not in payload or "pol_decision" not in payload:
      return reject("pol and pol_decision must both be present")
    if payload.pol_decision not in
       ["approved", "rejected", "pending_human_review"]:
      return reject("Invalid pol_decision value")

  // Validate DAG (against ledger or inline parent ECTs)
  result = validate_dag(payload, ledger,
                         clock_skew_tolerance)
  if result is error:
    return reject("DAG validation failed")

  // All checks passed; append to ledger
  ledger.append(payload)
  return accept

{: #fig-verification title="ECT Verification Pseudocode"}

Audit Ledger Interface

ECTs are designed to be recorded in an immutable audit ledger for compliance verification and post-hoc analysis. This specification defines required properties for the ledger but does not mandate a specific storage technology. Implementations MAY use append-only logs, databases with cryptographic commitment schemes, distributed ledgers, or any storage mechanism that provides the required properties.

An audit ledger implementation MUST provide:

Append-only semantics: Once an ECT is recorded, it MUST NOT be modified or deleted.
Ordering: The ledger MUST maintain a total ordering of ECT entries via a monotonically increasing sequence number.
Lookup by ECT ID: The ledger MUST support efficient retrieval of ECT entries by "jti" value.
Integrity verification: The ledger SHOULD provide a mechanism to verify that no entries have been tampered with (e.g., hash chains or Merkle trees).

The ledger SHOULD be maintained by an entity independent of the workflow agents to reduce the risk of collusion.

Use Cases

This section describes representative use cases demonstrating how ECTs provide execution records in regulated environments. These examples demonstrate ECT mechanics; production deployments would include additional domain-specific requirements beyond the scope of this specification.

Note: task identifiers in this section are abbreviated for readability. In production, all "jti" values are required to be UUIDs per {{exec-claims}}.

Medical Device SDLC Workflow

In a medical device software development lifecycle (SDLC), AI agents assist across multiple phases from requirements analysis through release approval. Regulatory frameworks including {{FDA-21CFR11}} Section 11.10(e) and {{EU-MDR}} require audit trails documenting the complete development process for software used in medical devices.

Agent A (Spec Reviewer):
  jti: task-001    par: []
  exec_act: review_requirements_spec
  pol: spec_review_policy_v2     pol_decision: approved

Agent B (Code Generator):
  jti: task-002    par: [task-001]
  exec_act: implement_module
  pol: coding_standards_v3       pol_decision: approved

Agent C (Test Agent):
  jti: task-003    par: [task-002]
  exec_act: execute_test_suite
  pol: test_coverage_policy_v1   pol_decision: approved

Agent D (Build Agent):
  jti: task-004    par: [task-003]
  exec_act: build_release_artifact
  pol: build_validation_v2       pol_decision: approved

Human Release Manager:
  jti: task-005    par: [task-004]
  exec_act: approve_release
  pol: release_approval_policy   pol_decision: approved
  pol_enforcer: spiffe://meddev.example/human/release-mgr-42
  ext: {org.ietf.wimse.witnessed_by: [...]}  (extension metadata)

{: #fig-medtech-sdlc title="Medical Device SDLC Workflow"}

ECTs record that requirements were reviewed before implementation began, that tests were executed against the implemented code, that the build artifact was validated, and that a human release manager explicitly approved the release. The DAG structure ensures no phase was skipped or reordered.

FDA Audit with DAG Reconstruction

During a regulatory audit, an FDA reviewer requests evidence of the development process for a specific software release. The auditing authority retrieves all ECTs sharing the same workflow identifier ("wid") from the audit ledger and reconstructs the complete DAG:

task-001 (review_requirements_spec)
  |
  v
task-002 (implement_module)
  |
  v
task-003 (execute_test_suite)
  |
  v
task-004 (build_release_artifact)
  |
  v
task-005 (approve_release) [human, witnessed]

{: #fig-fda-audit title="Reconstructed DAG for FDA Audit"}

The reconstructed DAG provides cryptographic evidence that:

Each phase was executed by an identified and authenticated agent.
Policy checkpoints were evaluated at every phase transition.
The execution sequence was maintained (no step was bypassed).
A human-in-the-loop approved the final release, with independent witness attestation.
Timestamps and execution durations are recorded for each step.

This can contribute to compliance with:

{{FDA-21CFR11}} Section 11.10(e): Computer-generated audit trails that record the date, time, and identity of the operator.
{{EU-MDR}} Annex II: Technical documentation traceability for the software development lifecycle.
{{EU-AI-ACT}} Article 12: Automatic logging capabilities for high-risk AI systems involved in the development process.
{{EU-AI-ACT}} Article 14: ECTs can record evidence that human oversight events occurred during the release process.

Financial Trading Workflow

In a financial trading workflow, agents perform risk assessment, compliance verification, and trade execution. The DAG structure records that compliance checks were evaluated before trade execution.

Agent A (Risk Assessment):
  jti: task-001    par: []
  exec_act: calculate_risk_exposure
  pol: risk_limits_policy_v2  pol_decision: approved

Agent B (Compliance):
  jti: task-002    par: [task-001]
  exec_act: verify_compliance
  pol: compliance_check_v1    pol_decision: approved

Agent C (Execution):
  jti: task-003    par: [task-002]
  exec_act: execute_trade
  pol: execution_policy_v3    pol_decision: approved

{: #fig-finance title="Financial Trading Workflow"}

This can contribute to compliance with:

{{MIFID-II}}: ECTs provide cryptographic records of the execution sequence that can support transaction audit requirements.
{{DORA}} Article 12: ECTs contribute to ICT activity logging.
{{EU-AI-ACT}} Article 12: Logging of decisions made by AI-driven systems.

Compensation and Rollback

When a compliance violation is discovered after execution, ECTs provide a mechanism to record authorized compensation actions with a cryptographic link to the original task:

{
  "iss": "spiffe://bank.example/agent/operations",
  "sub": "spiffe://bank.example/agent/operations",
  "aud": "spiffe://bank.example/system/ledger",
  "iat": 1772150550,
  "exp": 1772151150,
  "jti": "550e8400-e29b-41d4-a716-446655440099",
  "wid": "d3e4f5a6-b7c8-9012-def0-123456789012",
  "exec_act": "initiate_trade_rollback",
  "par": ["550e8400-e29b-41d4-a716-446655440003"],
  "pol": "compensation_policy_v1",
  "pol_decision": "approved",
  "pol_enforcer": "spiffe://bank.example/human/compliance-officer",
  "compensation_required": true,
  "compensation_reason": "policy_violation_in_parent_trade"
}

{: #fig-compensation title="Compensation ECT Example"}

The "par" claim links the compensation action to the original trade, creating an auditable chain from execution through violation discovery to remediation.

Autonomous Logistics Coordination

In a logistics workflow, multiple compliance checks complete before shipment commitment. The DAG structure records that all required checks were completed:

Agent A (Route Planning):
  jti: task-001    par: []
  exec_act: plan_route
  pol: route_policy_v1        pol_decision: approved

Agent B (Customs):
  jti: task-002    par: [task-001]
  exec_act: validate_customs
  pol: customs_policy_v2      pol_decision: approved

Agent C (Safety):
  jti: task-003    par: [task-001]
  exec_act: verify_cargo_safety
  pol: safety_policy_v1       pol_decision: approved

Agent D (Payment):
  jti: task-004    par: [task-002, task-003]
  exec_act: authorize_payment
  pol: payment_policy_v3      pol_decision: approved

System (Commitment):
  jti: task-005    par: [task-004]
  exec_act: commit_shipment
  pol: commitment_policy_v1   pol_decision: approved

{: #fig-logistics title="Logistics Workflow with Parallel Tasks"}

Note that tasks 002 and 003 both depend only on task-001 and can execute in parallel. Task 004 depends on both, demonstrating the DAG's ability to represent parallel execution with a join point.

Security Considerations

This section addresses security considerations following the guidance in {{RFC3552}}.

Threat Model

The following threat actors are considered:

Malicious agent (insider threat): An agent within the trust domain that intentionally creates false ECT claims.
Compromised agent (external attacker): An agent whose private key has been obtained by an external attacker.
Ledger tamperer: An entity attempting to modify or delete ledger entries after they have been recorded.
Time manipulator: An entity attempting to manipulate timestamps to alter perceived execution ordering.

Self-Assertion Limitation

ECTs are self-asserted by the executing agent. The agent claims what it did, and this claim is signed with its private key. A compromised or malicious agent could create ECTs with false claims (e.g., setting "pol_decision" to "approved" without actually evaluating the policy).

ECTs do not independently verify that:

The claimed execution actually occurred as described
The policy evaluation was correctly performed
The input/output hashes correspond to the actual data processed
The agent faithfully performed the stated action

The trustworthiness of ECT claims depends on the trustworthiness of the signing agent. To mitigate single-agent false claims, regulated environments SHOULD use the "org.ietf.wimse.witnessed_by" extension key (carried in "ext") to include independent third-party observers at critical decision points. However, this value is self-asserted by the ECT issuer: the listed witnesses do not co-sign the ECT and there is no cryptographic evidence within a single ECT that the witnesses actually observed the task. An issuing agent could list witnesses that did not participate.

Witness Attestation Model

To address the self-assertion limitation of the "org.ietf.wimse.witnessed_by" extension, witnesses SHOULD submit their own independent signed ECTs to the audit ledger attesting to the observed task. A witness attestation ECT:

MUST set "iss" to the witness's own workload identity.
MUST set "exec_act" to "witness_attestation" (or a domain- specific equivalent).
MUST include the observed task's "jti" in the "par" array, linking the attestation to the original task.
MUST set "pol_decision" to "approved" to indicate the witness confirms the observation.

When a task's "org.ietf.wimse.witnessed_by" extension lists one or more witnesses, auditors SHOULD verify that corresponding witness attestation ECTs exist in the ledger for each listed witness. A mismatch between the extension value and the set of independent witness ECTs in the ledger SHOULD be flagged during audit review.

This model converts witness attestation from a self-asserted claim to a cryptographically verifiable property of the ledger: the witness independently signs their own ECT using their own key, and the ledger records both the original task ECT and the witness attestation ECTs.

Organizational Prerequisites

ECTs operate within a broader trust framework. The guarantees provided by ECTs are only meaningful when the following organizational controls are in place:

Key management governance: Controls over who provisions agent keys and how keys are protected.
Ledger integrity governance: The ledger is maintained by an entity independent of the workflow agents.
Policy lifecycle management: Policy identifiers in ECTs map to actual, validated policy rules.
Agent deployment governance: Agents are deployed and maintained in a manner that preserves their integrity.

Signature Verification

ECTs MUST be signed with the agent's private key using JWS {{RFC7515}}. The signature algorithm MUST match the algorithm specified in the agent's WIT. Receivers MUST verify the ECT signature against the WIT public key before processing any claims. Receivers MUST verify that the signing key has not been revoked within the trust domain (see step 6 in {{verification}}).

If signature verification fails or if the signing key has been revoked, the ECT MUST be rejected entirely and the failure MUST be logged.

Implementations MUST use established JWS libraries and MUST NOT implement custom signature verification.

Replay Attack Prevention

ECTs include short expiration times (RECOMMENDED: 5-15 minutes) to limit the window for replay attacks. The "aud" claim restricts replay to unintended recipients: an ECT intended for Agent B will be rejected by Agent C. The "iat" claim enables receivers to reject ECTs that are too old, even if not yet expired.

The DAG structure provides additional replay protection: an ECT referencing parent tasks that already have a recorded child task with the same action can be flagged as a potential replay.

Implementations MUST maintain a cache of recently-seen "jti" values to detect replayed ECTs within the expiration window. An ECT with a duplicate "jti" value MUST be rejected.

Man-in-the-Middle Protection

ECTs do not replace transport-layer security. ECTs MUST be transmitted over TLS or mTLS connections. When used with the WIMSE service-to-service protocol {{I-D.ietf-wimse-s2s-protocol}}, transport security is already established. HTTP Message Signatures {{RFC9421}} provide an alternative channel binding mechanism.

The defense-in-depth model provides:

TLS/mTLS (transport layer): Prevents network-level tampering.
WIT/WPT (WIMSE identity layer): Proves agent identity and request authorization.
ECT (execution accountability layer): Records what the agent did and under what policy.

Key Compromise

If an agent's private key is compromised, an attacker can forge ECTs that appear to originate from that agent. To mitigate this risk:

Implementations SHOULD use short-lived keys and rotate them frequently (hours to days, not months).
Private keys SHOULD be stored in Hardware Security Modules (HSMs) or equivalent secure key storage.
Trust domains MUST support rapid key revocation.
Upon suspected compromise, the trust domain MUST revoke the compromised key and issue a new WIT with a fresh key pair.

ECTs signed with a compromised key that were recorded in the ledger before revocation remain valid historical records but SHOULD be flagged in the ledger as "signed with subsequently revoked key" for audit purposes.

Collusion and False Claims

A single malicious agent cannot forge parent task references because DAG validation requires parent tasks to exist in the ledger. However, multiple colluding agents could potentially create a false execution history if they control the ledger.

Mitigations include:

Independent ledger maintenance: The ledger SHOULD be maintained by an entity independent of the workflow agents.
Witness attestation: Using the "org.ietf.wimse.witnessed_by" extension key in "ext" to include independent third-party observers.
Cross-verification: Multiple independent ledger replicas can be compared for consistency.
Out-of-band audit: External auditors periodically verify ledger contents against expected workflow patterns.

Denial of Service

ECT signature verification is computationally inexpensive (approximately 1ms per ECT on modern hardware for ES256). DAG validation complexity is O(V) where V is the number of ancestor nodes reachable from the parent references; for typical shallow DAGs this is efficient.

Implementations SHOULD apply rate limiting at the API layer to prevent excessive ECT submissions. DAG validation SHOULD be performed after signature verification to avoid wasting resources on unsigned or incorrectly signed tokens.

Timestamp Accuracy

ECTs rely on timestamps ("iat", "exp") for temporal ordering. Clock skew between agents can lead to incorrect ordering judgments. Implementations SHOULD use synchronized time sources (e.g., NTP) and SHOULD allow a configurable clock skew tolerance (RECOMMENDED: 30 seconds).

Cross-organizational deployments where agents span multiple trust domains with independent time sources MAY require a higher clock skew tolerance. Deployments using trust domain federation SHOULD document their configured clock skew tolerance value and SHOULD ensure all participating trust domains agree on a common tolerance.

The temporal ordering check in DAG validation incorporates the clock skew tolerance to account for minor clock differences between agents.

ECT Size Constraints

ECTs with many parent tasks or large extension objects can increase HTTP header size. Implementations SHOULD limit the "par" array to a maximum of 256 entries. Workflows requiring more parent references SHOULD introduce intermediate aggregation tasks. The "ext" object SHOULD NOT exceed 4096 bytes when serialized as JSON and SHOULD NOT exceed a nesting depth of 5 levels (see also {{extension-claims}}).

Privacy Considerations

Data Exposure in ECTs

ECTs necessarily reveal:

Agent identities ("iss", "aud") for accountability purposes
Action descriptions ("exec_act") for audit trail completeness
Policy evaluation outcomes ("pol", "pol_decision") for compliance verification
Timestamps ("iat", "exp") for temporal ordering

ECTs are designed to NOT reveal:

Actual input or output data values (replaced with cryptographic hashes via "inp_hash" and "out_hash")
Internal computation details or intermediate steps
Proprietary algorithms or intellectual property
Personally identifiable information (PII)

Data Minimization

Implementations SHOULD minimize the information included in ECTs. The "exec_act" claim SHOULD use structured identifiers (e.g., "process_payment") rather than natural language descriptions. The "pol" claim SHOULD reference policy identifiers rather than embedding policy content.

The "compensation_reason" claim ({{compensation-claims}}) deserves particular attention: because it is human-readable and may describe the circumstances of a failure or policy violation, it risks exposing sensitive operational details. Implementations SHOULD use short, structured reason codes (e.g., "policy_violation_in_parent_trade") rather than free-form natural language explanations. Implementers SHOULD review "compensation_reason" values for potential information leakage before deploying to production.

Storage and Access Control

ECTs stored in audit ledgers SHOULD be access-controlled so that only authorized auditors and regulators can read them. Implementations SHOULD consider encryption at rest for ledger storage containing sensitive regulatory data.

Full input and output data (corresponding to the hashes in ECTs) SHOULD be stored separately from the ledger with additional access controls, since auditors may need to verify hash correctness but general access to the data values is not needed.

Regulatory Access

ECTs are designed for interpretation by qualified human auditors and regulators. ECTs provide structural records of execution ordering and policy evaluation; they are not intended for public disclosure.

IANA Considerations

Media Type Registration

This document requests registration of the following media type in the "Media Types" registry maintained by IANA:

Type name:: application
Subtype name:: wimse-exec+jwt
Required parameters:: none
Optional parameters:: none
Encoding considerations:: 8bit; an ECT is a JWT that is a JWS using the Compact Serialization, which is a sequence of Base64url-encoded values separated by period characters.
Security considerations:: See the Security Considerations section of this document.
Interoperability considerations:: none
Published specification:: This document
Applications that use this media type:: Applications that implement regulated agentic workflows requiring execution context tracing and audit trails.
Additional information:: Magic number(s): none File extension(s): none Macintosh file type code(s): none
Person and email address to contact for further information:: Christian Nennemann, ietf@nennemann.de
Intended usage:: COMMON
Restrictions on usage:: none
Author:: Christian Nennemann
Change controller:: IETF

HTTP Header Field Registration

This document requests registration of the following header field in the "Hypertext Transfer Protocol (HTTP) Field Name Registry" maintained by IANA:

Field name:: Execution-Context
Status:: permanent
Specification document:: This document, {{http-header}}

JWT Claims Registration

This document requests registration of the following claims in the "JSON Web Token Claims" registry maintained by IANA:

Claim Name	Claim Description	Change Controller	Reference
wid	Workflow Identifier	IETF	{{exec-claims}}
exec_act	Action/Task Type	IETF	{{exec-claims}}
par	Parent Task Identifiers	IETF	{{exec-claims}}
pol	Policy Rule Identifier	IETF	{{policy-claims}}
pol_decision	Policy Decision Result	IETF	{{policy-claims}}
pol_enforcer	Policy Enforcer Identity	IETF	{{policy-claims}}
inp_hash	Input Data Hash	IETF	{{data-integrity-claims}}
out_hash	Output Data Hash	IETF	{{data-integrity-claims}}
compensation_required	Compensation Flag	IETF	{{compensation-claims}}
compensation_reason	Compensation Reason	IETF	{{compensation-claims}}
ext	Extension Object	IETF	{{extension-claims}}
{: #table-claims title="JWT Claims Registrations"}

ECT Policy Decision Values Registry

This document establishes the "ECT Policy Decision Values" registry under the "JSON Web Token (JWT)" group. Registration policy is Specification Required per {{!RFC8126}}.

The initial contents of the registry are:

Value	Description	Change Controller	Reference
approved	Policy evaluation succeeded	IETF	{{policy-claims}}
rejected	Policy evaluation failed	IETF	{{policy-claims}}
pending_human_review	Awaiting human judgment	IETF	{{policy-claims}}
{: #table-pol-decision title="ECT Policy Decision Values"}

--- back

{:numbered="false"}

WIMSE Workload Identity

{:numbered="false"}

The WIMSE architecture {{I-D.ietf-wimse-arch}} and service-to- service protocol {{I-D.ietf-wimse-s2s-protocol}} provide the identity foundation upon which ECTs are built. WIT/WPT answer "who is this agent?" and "does it control the claimed key?" while ECTs record "what did this agent do?" and "what policy was evaluated?" Together they form an identity-plus-accountability framework for regulated agentic systems.

OAuth 2.0 Token Exchange and the "act" Claim

{:numbered="false"}

{{RFC8693}} defines the OAuth 2.0 Token Exchange protocol and registers the "act" (Actor) claim in the JWT Claims registry. The "act" claim creates nested JSON objects representing a delegation chain: "who is acting on behalf of whom." While the nesting superficially resembles a chain, it is strictly linear (each "act" object contains at most one nested "act"), represents authorization delegation rather than task execution, and carries no task identifiers, policy decisions, or input/output integrity data. The "act" chain cannot represent branching (fan-out) or convergence (fan-in) and therefore cannot form a DAG.

ECTs intentionally use the distinct claim name "exec_act" for the action/task type to avoid collision with the "act" claim. The two concepts are orthogonal: "act" records "who authorized whom," ECTs record "what was done, in what order, with what policy outcomes."

Transaction Tokens

{:numbered="false"}

OAuth Transaction Tokens {{I-D.ietf-oauth-transaction-tokens}} propagate authorization context across workload call chains. The Txn-Token "req_wl" claim accumulates a comma-separated list of workloads that requested replacement tokens, which is the closest existing mechanism to call-chain recording.

However, "req_wl" cannot form a DAG because:

It is linear: a comma-separated string with no branching or merging representation. When a workload fans out to multiple downstream services, each receives the same "req_wl" value and the branching is invisible.
It is incomplete: only workloads that request a replacement token from the Transaction Token Service appear in "req_wl"; workloads that forward the token unchanged are not recorded.
It carries no task-level granularity, no parent references, no policy evaluation outcomes, and no execution content.

Extensions for agentic use cases ({{I-D.oauth-transaction-tokens-for-agents}}) add agent identity and constraints ("agentic_ctx") but no execution ordering or DAG structure.

ECTs and Transaction Tokens are complementary: a Txn-Token propagates authorization context ("this request is authorized for scope X on behalf of user Y"), while an ECT records execution accountability ("task T was performed, depending on tasks P1 and P2, with policy Z evaluated and approved"). An agent request could carry both a Txn-Token for authorization and an ECT for execution recording. The WPT "tth" claim defined in {{I-D.ietf-wimse-s2s-protocol}} can hash-bind a WPT to a co-present Txn-Token; a similar binding mechanism for ECTs is a potential future extension.

Distributed Tracing (OpenTelemetry)

{:numbered="false"}

OpenTelemetry {{OPENTELEMETRY}} and similar distributed tracing systems provide observability for debugging and monitoring. ECTs differ in several important ways: ECTs are cryptographically signed per-task with the agent's private key; ECTs are tamper-evident through JWS signatures; ECTs enforce DAG validation rules; and ECTs are designed for regulatory audit rather than operational monitoring. OpenTelemetry data is typically controlled by the platform operator and can be modified or deleted without detection. ECTs and distributed traces are complementary: traces provide observability while ECTs provide signed execution records. ECTs may reference OpenTelemetry trace identifiers in the "ext" claim for correlation.

Blockchain and Distributed Ledgers

{:numbered="false"}

Both ECTs and blockchain systems provide immutable records. This specification intentionally defines only the ECT token format and is agnostic to the storage mechanism. ECTs can be stored in append-only logs, databases with cryptographic commitments, blockchain networks, or any storage providing the required properties defined in {{ledger-interface}}.

SCITT (Supply Chain Integrity, Transparency, and Trust)

{:numbered="false"}

The SCITT architecture {{I-D.ietf-scitt-architecture}} defines a framework for creating transparent and auditable supply chain records through Transparency Services, Signed Statements, and Receipts. ECTs and SCITT are naturally complementary: the ECT "wid" (Workflow Identifier) claim can serve as a correlation identifier referenced in SCITT Signed Statements, linking a complete ECT audit trail to a supply chain transparency record. For example, in a regulated manufacturing workflow, each agent step produces an ECT (recording what was done, by whom, under what policy), while the overall workflow identified by "wid" is registered as a SCITT Signed Statement on a Transparency Service. This enables auditors to verify both the individual execution steps (via ECT DAG validation) and the end-to-end supply chain integrity (via SCITT Receipts) using the "wid" as the shared correlation point. The "ext" claim in ECTs ({{exec-claims}}) can carry SCITT-specific metadata such as Transparency Service identifiers or Receipt references for tighter integration.

W3C Verifiable Credentials

{:numbered="false"}

W3C Verifiable Credentials represent claims about subjects (e.g., identity, qualifications). ECTs represent execution records of actions (what happened, in what order, under what policy). While both use JWT/JWS as a serialization format, their semantics and use cases are distinct.

Implementation Guidance

{:numbered="false"}

Minimal Implementation

{:numbered="false"}

A minimal conforming implementation needs to:

Create JWTs with all required claims ("iss", "aud", "iat", "exp", "jti", "exec_act", "par") and policy claims ("pol", "pol_decision") when policy evaluation was performed.
Sign ECTs with the agent's private key using an algorithm matching the WIT (ES256 recommended).
Verify ECT signatures against WIT public keys.
Perform DAG validation (parent existence, temporal ordering, cycle detection).
Append verified ECTs to the audit ledger.

Storage Recommendations

{:numbered="false"}

Append-only log: Simplest approach; immutability by design.
Database with hash chains: Periodic cryptographic commitments over batches of entries.
Distributed ledger: Maximum immutability guarantees for cross-organizational audit.
Hybrid: Hot storage in a database, cold archive in immutable storage.

Performance Considerations

{:numbered="false"}

ES256 signature verification: approximately 1ms per ECT on modern hardware.
DAG validation: O(V) where V is the number of reachable ancestor nodes (typically small for shallow workflows).
JSON serialization: sub-millisecond per ECT.
Total per-request overhead: approximately 5-10ms, acceptable for regulated workflows where correctness is prioritized over latency.

Interoperability

{:numbered="false"}

Implementations are expected to use established JWT/JWS libraries (JOSE) for token creation and verification. Custom cryptographic implementations are strongly discouraged. Implementations are expected to be tested against multiple JWT libraries to ensure interoperability.

Regulatory Compliance Mapping

{:numbered="false"}

The following table summarizes how ECTs can contribute to compliance with various regulatory frameworks. ECTs are a technical building block; achieving compliance requires additional organizational measures beyond this specification.

Regulation	Requirement	ECT Contribution
FDA 21 CFR Part 11	Audit trails recording date, time, operator, actions (11.10(e))	Cryptographic signatures and append-only ledger contribute to audit trail requirements
EU MDR	Technical documentation traceability (Annex II)	ECTs provide signed records of AI-assisted decision sequences
EU AI Act Art. 12	Automatic logging capabilities for high-risk AI	ECTs contribute cryptographic activity logging
EU AI Act Art. 14	Human oversight capability	ECTs can record evidence that human oversight events occurred
MiFID II	Transaction records for supervisory authorities	ECTs provide cryptographic execution sequence records
DORA Art. 12	ICT activity logging policies	ECT ledger contributes to ICT activity audit trail
{: #table-regulatory title="Regulatory Compliance Mapping"}

Examples

{:numbered="false"}

Example 1: Simple Two-Agent Workflow

{:numbered="false"}

Agent A executes a data retrieval task and sends the ECT to Agent B:

ECT JOSE Header:

{
  "alg": "ES256",
  "typ": "wimse-exec+jwt",
  "kid": "agent-a-key-2026-02"
}

ECT Payload:

{
  "iss": "spiffe://example.com/agent/data-retrieval",
  "sub": "spiffe://example.com/agent/data-retrieval",
  "aud": "spiffe://example.com/agent/validator",
  "iat": 1772064150,
  "exp": 1772064750,
  "jti": "550e8400-e29b-41d4-a716-446655440001",
  "wid": "b1c2d3e4-f5a6-7890-bcde-f01234567890",
  "exec_act": "fetch_patient_data",
  "par": [],
  "pol": "clinical_data_access_policy_v1",
  "pol_decision": "approved",
  "inp_hash": "sha-256:n4bQgYhMfWWaL-qgxVrQFaO_TxsrC4Is0V1sFbDwCgg",
  "out_hash": "sha-256:LCa0a2j_xo_5m0U8HTBBNBNCLXBkg7-g-YpeiGJm564"
}

Agent B receives the ECT, verifies it, executes a validation task, and creates its own ECT:

{
  "iss": "spiffe://example.com/agent/validator",
  "sub": "spiffe://example.com/agent/validator",
  "aud": "spiffe://example.com/system/ledger",
  "iat": 1772064160,
  "exp": 1772064760,
  "jti": "550e8400-e29b-41d4-a716-446655440002",
  "wid": "b1c2d3e4-f5a6-7890-bcde-f01234567890",
  "exec_act": "validate_safety",
  "par": ["550e8400-e29b-41d4-a716-446655440001"],
  "pol": "safety_validation_policy_v2",
  "pol_decision": "approved"
}

The resulting DAG:

task-...-0001 (fetch_patient_data)
  |
  v
task-...-0002 (validate_safety)

Example 2: Medical Device SDLC with Release Approval

{:numbered="false"}

A multi-step medical device software lifecycle workflow with autonomous agents and human release approval:

Task 1 (Spec Review Agent):

{
  "iss": "spiffe://meddev.example/agent/spec-reviewer",
  "sub": "spiffe://meddev.example/agent/spec-reviewer",
  "aud": "spiffe://meddev.example/agent/code-gen",
  "iat": 1772064150,
  "exp": 1772064750,
  "jti": "a1b2c3d4-0001-0000-0000-000000000001",
  "wid": "c2d3e4f5-a6b7-8901-cdef-012345678901",
  "exec_act": "review_requirements_spec",
  "par": [],
  "pol": "spec_review_policy_v2",
  "pol_decision": "approved",
  "inp_hash": "sha-256:n4bQgYhMfWWaL-qgxVrQFaO_TxsrC4Is0V1sFbDwCgg",
  "out_hash": "sha-256:LCa0a2j_xo_5m0U8HTBBNBNCLXBkg7-g-YpeiGJm564"
}

Task 2 (Code Generation Agent):

{
  "iss": "spiffe://meddev.example/agent/code-gen",
  "sub": "spiffe://meddev.example/agent/code-gen",
  "aud": "spiffe://meddev.example/agent/test-runner",
  "iat": 1772064200,
  "exp": 1772064800,
  "jti": "a1b2c3d4-0001-0000-0000-000000000002",
  "wid": "c2d3e4f5-a6b7-8901-cdef-012345678901",
  "exec_act": "implement_module",
  "par": ["a1b2c3d4-0001-0000-0000-000000000001"],
  "pol": "coding_standards_v3",
  "pol_decision": "approved"
}

Task 3 (Autonomous Test Agent):

{
  "iss": "spiffe://meddev.example/agent/test-runner",
  "sub": "spiffe://meddev.example/agent/test-runner",
  "aud": "spiffe://meddev.example/agent/build",
  "iat": 1772064260,
  "exp": 1772064860,
  "jti": "a1b2c3d4-0001-0000-0000-000000000003",
  "wid": "c2d3e4f5-a6b7-8901-cdef-012345678901",
  "exec_act": "execute_test_suite",
  "par": ["a1b2c3d4-0001-0000-0000-000000000002"],
  "pol": "test_coverage_policy_v1",
  "pol_decision": "approved"
}

Task 4 (Build Agent):

{
  "iss": "spiffe://meddev.example/agent/build",
  "sub": "spiffe://meddev.example/agent/build",
  "aud": "spiffe://meddev.example/human/release-mgr-42",
  "iat": 1772064310,
  "exp": 1772064910,
  "jti": "a1b2c3d4-0001-0000-0000-000000000004",
  "wid": "c2d3e4f5-a6b7-8901-cdef-012345678901",
  "exec_act": "build_release_artifact",
  "par": ["a1b2c3d4-0001-0000-0000-000000000003"],
  "pol": "build_validation_v2",
  "pol_decision": "approved",
  "out_hash": "sha-256:Ry1YfOoW2XpC5Mq8HkGzNx3dL9vBa4sUjE7iKt0wPZc"
}

Task 5 (Human Release Manager Approval):

{
  "iss": "spiffe://meddev.example/human/release-mgr-42",
  "sub": "spiffe://meddev.example/human/release-mgr-42",
  "aud": "spiffe://meddev.example/system/ledger",
  "iat": 1772064510,
  "exp": 1772065110,
  "jti": "a1b2c3d4-0001-0000-0000-000000000005",
  "wid": "c2d3e4f5-a6b7-8901-cdef-012345678901",
  "exec_act": "approve_release",
  "par": ["a1b2c3d4-0001-0000-0000-000000000004"],
  "pol": "release_approval_policy",
  "pol_decision": "approved",
  "pol_enforcer": "spiffe://meddev.example/human/release-mgr-42",
  "ext": {
    "org.ietf.wimse.witnessed_by": [
      "spiffe://meddev.example/audit/qa-observer-1"
    ]
  }
}

The resulting DAG records the complete SDLC: spec review preceded implementation, implementation preceded testing, testing preceded build, and a human release manager approved the final release. The "ext" object in task 5 carries witness metadata via the "org.ietf.wimse.witnessed_by" extension key.

task-...-0001 (review_requirements_spec)
  |
  v
task-...-0002 (implement_module)
  |
  v
task-...-0003 (execute_test_suite)
  |
  v
task-...-0004 (build_release_artifact)
  |
  v
task-...-0005 (approve_release) [human]

An FDA auditor reconstructs this DAG by querying the audit ledger for all ECTs with wid "c2d3e4f5-a6b7-8901-cdef-012345678901" and verifying each signature. The DAG provides cryptographic evidence that the SDLC followed the prescribed process with human oversight at the release gate.

Example 3: Parallel Execution with Join

{:numbered="false"}

A workflow where two tasks execute in parallel and a third task depends on both:

task-...-0001 (assess_risk)
  |            \
  v             v
task-...-0002  task-...-0003
(check         (verify
compliance)    liquidity)
  |            /
  v           v
task-...-0004 (execute_trade)

Task 004 ECT payload:

{
  "iss": "spiffe://bank.example/agent/execution",
  "sub": "spiffe://bank.example/agent/execution",
  "aud": "spiffe://bank.example/system/ledger",
  "iat": 1772064250,
  "exp": 1772064850,
  "jti": "f1e2d3c4-0004-0000-0000-000000000004",
  "wid": "d3e4f5a6-b7c8-9012-def0-123456789012",
  "exec_act": "execute_trade",
  "par": [
    "f1e2d3c4-0002-0000-0000-000000000002",
    "f1e2d3c4-0003-0000-0000-000000000003"
  ],
  "pol": "trade_execution_policy_v3",
  "pol_decision": "approved"
}

The "par" array with two entries records that both compliance checking and liquidity verification were completed before trade execution.

Acknowledgments

{:numbered="false"}

The author thanks the WIMSE working group for their foundational work on workload identity in multi-system environments. The concepts of Workload Identity Tokens and Workload Proof Tokens provide the identity foundation upon which execution context tracing is built.

72 KiB Raw Blame History

Introduction

Motivation

Problem Statement

Scope and Applicability

Relationship to Regulatory Compliance

Conventions and Definitions

WIMSE Architecture Integration

WIMSE Foundation

Extension Model

Integration Points

Execution Context Token Format

JOSE Header

JWT Claims

Standard JWT Claims

Execution Context

Policy Evaluation

Data Integrity

Compensation and Rollback

Extensions

Complete ECT Example

HTTP Header Transport

Execution-Context Header Field

DAG Validation

Overview

Validation Rules

DAG Validation Algorithm

Signature and Token Verification

Verification Procedure

Verification Pseudocode

Audit Ledger Interface

Use Cases

Medical Device SDLC Workflow

FDA Audit with DAG Reconstruction

Financial Trading Workflow

Compensation and Rollback

Autonomous Logistics Coordination

Security Considerations

Threat Model

Self-Assertion Limitation

Witness Attestation Model

Organizational Prerequisites

Signature Verification

Replay Attack Prevention

Man-in-the-Middle Protection

Key Compromise

Collusion and False Claims

Denial of Service

Timestamp Accuracy

ECT Size Constraints

Privacy Considerations

Data Exposure in ECTs

Data Minimization

Storage and Access Control

Regulatory Access

IANA Considerations

Media Type Registration

HTTP Header Field Registration

JWT Claims Registration

ECT Policy Decision Values Registry

Related Work

WIMSE Workload Identity

OAuth 2.0 Token Exchange and the "act" Claim

Transaction Tokens

Distributed Tracing (OpenTelemetry)

Blockchain and Distributed Ledgers

SCITT (Supply Chain Integrity, Transparency, and Trust)

W3C Verifiable Credentials

Implementation Guidance

Minimal Implementation

Storage Recommendations

Performance Considerations

Interoperability

Regulatory Compliance Mapping

Examples

Example 1: Simple Two-Agent Workflow

Example 2: Medical Device SDLC with Release Approval

Example 3: Parallel Execution with Join

Acknowledgments

72 KiB

Raw Blame History