{
  "schemaVersion": 1,
  "id": "atlas-lesson-delegated-agent-protocols",
  "title": "Delegated tasks, messages and artifacts",
  "category": "learning",
  "format": "Atlas lesson",
  "description": "Task lifecycle, artifact ownership, input-required states, cancellation and authenticated service boundaries.",
  "canonicalUrl": "https://semiaifoundry.com/learn/atlas/delegated-agent-protocols/",
  "publishedDate": null,
  "revisionDate": "2026-10-07",
  "author": null,
  "topic": "Delegated tasks and coordination",
  "series": null,
  "license": null,
  "rightsUrl": "https://semiaifoundry.com/ai-atlas/licenses.html",
  "version": "1.3.0",
  "status": "Published educational lesson",
  "evidenceScope": "This is a worked explanation, not an executing external platform, model or robot. The quantities and fixtures are authored. Source scope is recorded in the automation learning review manifest; no source benchmark or private system behavior is claimed. Worked lifecycle based on the inspected A2A specification; no network endpoint, interoperability test or multi-agent runtime is executed. The concrete protocol revision must be pinned for a future adapter.",
  "contentCoverage": "published-lesson",
  "artifactUrl": "https://semiaifoundry.com/ai-atlas/lesson-data/delegated-agent-protocols-f7d8f254309a16f8.json",
  "interactiveUrl": "https://semiaifoundry.com/ai-atlas/#delegated-agent-protocols",
  "related": [],
  "supersedes": [],
  "supersededBy": [],
  "contentHash": "3262105dca75bcf8d24d2163ade24f792ba59c3ae3b89e8b9386e02f2e967183",
  "contentLength": 6416,
  "jsonUrl": "https://semiaifoundry.com/agents/publications/atlas-lesson-delegated-agent-protocols.json",
  "markdownUrl": "https://semiaifoundry.com/agents/publications/atlas-lesson-delegated-agent-protocols.md",
  "sections": [
    {
      "id": "section-1",
      "title": "Delegated tasks, messages and artifacts",
      "start": 49,
      "end": 508
    },
    {
      "id": "section-2",
      "title": "The mechanism",
      "start": 508,
      "end": 1774
    },
    {
      "id": "section-3",
      "title": "Science and assumptions",
      "start": 1774,
      "end": 1802
    },
    {
      "id": "section-4",
      "title": "Delegated tasks, messages and artifacts · explicit relationship",
      "start": 1802,
      "end": 2800
    },
    {
      "id": "section-5",
      "title": "Engineering example",
      "start": 2800,
      "end": 3348
    },
    {
      "id": "section-6",
      "title": "Distinctions and tradeoffs",
      "start": 3348,
      "end": 3727
    },
    {
      "id": "section-7",
      "title": "Specifications and interfaces",
      "start": 3727,
      "end": 4228
    },
    {
      "id": "section-8",
      "title": "Interfaces",
      "start": 4228,
      "end": 4534
    },
    {
      "id": "section-9",
      "title": "Check your understanding",
      "start": 4534,
      "end": 4563
    },
    {
      "id": "section-10",
      "title": "The network disconnects while delegated task T17 is working. What should the parent avoid?",
      "start": 4563,
      "end": 5001
    },
    {
      "id": "section-11",
      "title": "Evidence and scope",
      "start": 5001,
      "end": 5618
    },
    {
      "id": "section-12",
      "title": "Sources",
      "start": 5618,
      "end": 5970
    },
    {
      "id": "section-13",
      "title": "Continue exploring",
      "start": 5970,
      "end": 5993
    },
    {
      "id": "section-14",
      "title": "Foundations",
      "start": 5993,
      "end": 6226
    },
    {
      "id": "section-15",
      "title": "Related lessons",
      "start": 6226,
      "end": 6416
    }
  ],
  "citations": [
    {
      "title": "Explore in Atlas",
      "url": "https://semiaifoundry.com/ai-atlas/#delegated-agent-protocols"
    },
    {
      "title": "A2A specification",
      "url": "https://a2a-protocol.org/latest/specification/"
    },
    {
      "title": "Temporal Activities",
      "url": "https://docs.temporal.io/activities"
    },
    {
      "title": "Published lesson data",
      "url": "https://semiaifoundry.com/ai-atlas/lesson-data/delegated-agent-protocols-f7d8f254309a16f8.json"
    },
    {
      "title": "Atlas releases and licensing",
      "url": "https://semiaifoundry.com/ai-atlas/licenses.html"
    },
    {
      "title": "A tool protocol is a boundary",
      "url": "https://semiaifoundry.com/learn/atlas/mcp-host-client-server/"
    },
    {
      "title": "Durable actions and the ambiguous timeout",
      "url": "https://semiaifoundry.com/learn/atlas/durable-actions-idempotency/"
    },
    {
      "title": "Tools, memory & agent loop",
      "url": "https://semiaifoundry.com/learn/atlas/tool-loop/"
    },
    {
      "title": "Evaluation & the AGI frontier",
      "url": "https://semiaifoundry.com/learn/atlas/evaluation/"
    }
  ],
  "recordHash": "f0f5276253b4c31f7e40c5733b546f25a9d4b32320837595dbdc5e7473087dcf",
  "contentFormat": "text/markdown",
  "content": "Delegated tasks and coordination · Atlas v1.3.0\n\n# Delegated tasks, messages and artifacts\n\nTask lifecycle, artifact ownership, input-required states, cancellation and authenticated service boundaries.\n\n[Explore in Atlas](https://semiaifoundry.com/ai-atlas/#delegated-agent-protocols)Logical state, task evidence and declared units; no physical assembly scale\n\nLearning objectiveTrack task identity, partial artifacts and input-required states without equating remote completion with accepted task success.\n\n## The mechanism\n\nDelegation hands a bounded task to another participant and receives messages, artifacts and status over time. That interaction can outlive a single request. The parent must know which task is being updated, who owns its artifacts and whether it is working, waiting for input, completed, failed or canceled. A2A defines a task-oriented protocol surface for such exchanges; this does not prove that either participant is intelligent or correct. A tool call and a delegated task can both be asynchronous, but they may expose different lifecycles and ownership contracts. The delegator should not turn a dropped connection into a new task without checking the existing identity. It should also retain its own final-goal and authority checks rather than treating a remote completed status as permission to publish an artifact.\n\n1.  The request declares a task identity, goal and permitted information/effects for the receiving service.\n2.  Status and artifact updates refer to that identity while execution continues.\n3.  An input-required state asks for missing information without claiming completion or starting a duplicate task.\n4.  The parent verifies the returned artifact and applies its own task and permission contract before further effects.\n\n## Science and assumptions\n\n### Delegated tasks, messages and artifacts · explicit relationship\n\nA delegated task retains its identity across messages, input requests and artifact updates. Remote completion is checked against the parent’s task and authority before further effects.\n\nA fictional delegated review task T17 returns a partial inspection note while its state is working. It later requests the asset revision and remains input-required. Sending that revision resumes T17; creating unrelated T18 instead could duplicate work and lose the original context. When T17 reports completion with an artifact, the parent checks that it refers to FAN-7 and the requested revision. A completed artifact for FAN-9 remains wrong for this task. This is a worked message lifecycle, not a running A2A endpoint.\n\nAssumptions Worked lifecycle based on the inspected A2A specification; no network endpoint, interoperability test or multi-agent runtime is executed. The concrete protocol revision must be pinned for a future adapter.\n\n## Engineering example\n\nA fictional delegated review task T17 returns a partial inspection note while its state is working. It later requests the asset revision and remains input-required. Sending that revision resumes T17; creating unrelated T18 instead could duplicate work and lose the original context. When T17 reports completion with an artifact, the parent checks that it refers to FAN-7 and the requested revision. A completed artifact for FAN-9 remains wrong for this task. This is a worked message lifecycle, not a running A2A endpoint.\n\n## Distinctions and tradeoffs\n\nDelegation does not transfer unlimited authority or make a remote completion flag an independent correctness verdict.\n\nTask-oriented protocols support long-lived work but require explicit ownership, reconnection and failure handling.\n\nCoordinated AI systems need identities and outcome contracts between participants, not merely more agent names.\n\n## Specifications and interfaces\n\nMechanism boundary\n\nHow is an unfinished delegated task represented?\n\nWorked lifecycle based on the inspected A2A specification; no network endpoint, interoperability test or multi-agent runtime is executed. The concrete protocol revision must be pinned for a future adapter.\n\nEvidence type\n\nAuthored worked example\n\nWorked explanation with a bounded numeric or state example. This lesson does not claim an executing implementation of an external model or platform.\n\n### Interfaces\n\nDeclared input and state Input\n\nThe request declares a task identity, goal and permitted information/effects for the receiving service.\n\nChecked outcome Output\n\nTrack task identity, partial artifacts and input-required states without equating remote completion with accepted task success.\n\n## Check your understanding\n\n### The network disconnects while delegated task T17 is working. What should the parent avoid?\n\n1.  Checking T17’s existing status\n2.  Blindly creating a duplicate task for the same intended effect\n3.  Retaining the task identity\n\nAnswer and reasoning\n\nBlindly creating a duplicate task for the same intended effect\n\nA lost connection does not establish that the task stopped; duplication must be controlled at the task/effect boundary.\n\n## Evidence and scope\n\nThis is a worked explanation, not an executing external platform, model or robot. The quantities and fixtures are authored. Source scope is recorded in the automation learning review manifest; no source benchmark or private system behavior is claimed. Worked lifecycle based on the inspected A2A specification; no network endpoint, interoperability test or multi-agent runtime is executed. The concrete protocol revision must be pinned for a future adapter.\n\nThe worked example explains this mechanism. General numerical labs are a reference; they do not execute this lesson’s full mechanism.\n\n## Sources\n\n-   [A2A specification](https://a2a-protocol.org/latest/specification/)\n-   [Temporal Activities](https://docs.temporal.io/activities)\n\n[Published lesson data](https://semiaifoundry.com/ai-atlas/lesson-data/delegated-agent-protocols-f7d8f254309a16f8.json) · [Atlas releases and licensing](https://semiaifoundry.com/ai-atlas/licenses.html)\n\n## Continue exploring\n\n### Foundations\n\n-   [A tool protocol is a boundary](https://semiaifoundry.com/learn/atlas/mcp-host-client-server/)\n-   [Durable actions and the ambiguous timeout](https://semiaifoundry.com/learn/atlas/durable-actions-idempotency/)\n\n### Related lessons\n\n-   [Tools, memory & agent loop](https://semiaifoundry.com/learn/atlas/tool-loop/)\n-   [Evaluation & the AGI frontier](https://semiaifoundry.com/learn/atlas/evaluation/)"
}