{
  "schema": "vedokrok.public-item.v1",
  "release_id": "MHC-RPUB-20260920-75ad787a",
  "url": "/knowledge/run-agent-actions-in-the-user-s-authorization-context",
  "id": "MHC-D-RESEARCH-0311",
  "version": "0.1.0",
  "title": "Run agent actions in the user's authorization context",
  "summary": "A helpful assistant should not quietly become a shared administrator.",
  "kind": "principle",
  "body": "Carry the initiating user's identity and allowed scope into downstream calls instead of using one generic high-privilege service identity for everyone. Let the destination system enforce what that user can access. This keeps the agent from turning a reasoning mistake into cross-user authority.",
  "limits": [
    "User-context execution does not make malicious or mistaken user requests safe; policy and high-impact approval may still apply."
  ],
  "topics": [
    "union-ai-agent-control"
  ],
  "intents": [],
  "source_ids": [
    "RS-EEEBBCF69A614717"
  ],
  "evidence": [
    {
      "claim": "OWASP recommends executing agent extensions in the specific user's authorization context with the minimum necessary scope.",
      "source_id": "RS-EEEBBCF69A614717",
      "role": "supports",
      "note": "User-context execution still requires server-side authorization and does not make every requested action legitimate.",
      "locator": "Execute extensions in user's context"
    }
  ],
  "use_when": [
    "An AI acts on behalf of different users against shared downstream systems."
  ],
  "avoid_when": [
    "User-context execution does not make malicious or mistaken user requests safe; policy and high-impact approval may still apply."
  ],
  "example": "A repository assistant uses the user's OAuth scope for the selected repository rather than a token that can edit every repository in the organization.",
  "check": "Changing the user changes the resources and actions the downstream system authorizes.",
  "sources": [
    {
      "id": "RS-EEEBBCF69A614717",
      "title": "LLM06:2025 Excessive Agency",
      "url": "https://genai.owasp.org/llmrisk/llm062025-excessive-agency/"
    }
  ],
  "relations": [
    {
      "from": "MHC-D-RESEARCH-0311",
      "to": "MHC-D-RESEARCH-0315",
      "type": "useful_with",
      "url": "/knowledge/enforce-permissions-outside-the-model"
    }
  ],
  "collections": [
    {
      "id": "RC-EEF5F4530C11FD17",
      "title": "Let AI do useful work without giving it accidental authority",
      "url": "/collections/let-ai-do-useful-work-without-giving-it-accidental-authority"
    }
  ]
}