{
  "id": "human-approval-gates",
  "type": "entity",
  "name": "Human approval gates and graduated autonomy",
  "summary": "The control that keeps an agent's blast radius human-sized: specific action classes (destructive, irreversible, outward-facing) require a human yes before execution, and autonomy expands only as the agent earns it with evidence. The complement to sandboxing — the sandbox contains what the agent can do, gates contain what it may do.",
  "locale": "en",
  "tags": [
    "governance",
    "autonomy",
    "approval-gates",
    "oversight",
    "safety"
  ],
  "relations": [
    {
      "rel": "complements",
      "target": "sandboxed-execution"
    },
    {
      "rel": "mitigates",
      "target": "prompt-injection-testing"
    },
    {
      "rel": "related",
      "target": "agent-telemetry-actor-classification"
    }
  ],
  "questions": [
    "Which agent actions should require human approval?",
    "How do I decide how much autonomy to give an agent?",
    "What is graduated autonomy?"
  ],
  "claims": [
    {
      "id": "c1",
      "text": "Anthropic's agent-engineering guidance is explicit: agents should pause for human feedback at checkpoints or when encountering blockers, because autonomous operation compounds errors — which is why it pairs the advice with extensive sandboxed testing and guardrails.",
      "sources": [
        "anthropic-building-agents"
      ],
      "confidence": 0.9
    },
    {
      "id": "c2",
      "text": "OWASP ranks Excessive Agency (LLM06:2025) as a top-10 risk: an agent granted more autonomy or permissions than its task requires turns any upstream failure — injection, hallucination — into unauthorized actions.",
      "sources": [
        "owasp-llm-top10"
      ],
      "confidence": 0.9
    },
    {
      "id": "c3",
      "text": "Autonomy should be graduated, not granted: add agent complexity only when it demonstrably improves outcomes, starting from the simplest solution that works.",
      "sources": [
        "anthropic-building-agents"
      ],
      "confidence": 0.85
    },
    {
      "id": "c4",
      "text": "A production example of gates as configuration: this instance's own blueprint declares an autonomy level and named human gates (blueprint approval, ship approval) as machine-readable contract, with exceptions that expire after a fixed period.",
      "sources": [
        "citarium-methodology-brief"
      ],
      "confidence": 0.9
    }
  ],
  "takeaways": [],
  "faqs": [],
  "evidence_tier": "primary",
  "evidence": {
    "level": "production",
    "source_types": [
      "production_system",
      "personal_experience",
      "industry_observation"
    ]
  },
  "moat_flag": true,
  "winning_edge": "Frames gates and sandboxes as orthogonal controls (may-do vs can-do) and shows gates as declared, expiring configuration from a system that runs them — not as a policy paragraph nobody can test.",
  "confidence": 0.85,
  "last_verified": "2026-08-08",
  "canonical_url": "https://agentreliability.dev/k/human-approval-gates",
  "api_url": "https://agentreliability.dev/api/k/human-approval-gates.json",
  "jsonld": {
    "@context": "https://schema.org",
    "name": "Human approval gates and graduated autonomy",
    "description": "The control that keeps an agent's blast radius human-sized: specific action classes (destructive, irreversible, outward-facing) require a human yes before execution, and autonomy expands only as the agent earns it with evidence. The complement to sandboxing — the sandbox contains what the agent can do, gates contain what it may do.",
    "url": "https://agentreliability.dev/k/human-approval-gates",
    "license": "https://spdx.org/licenses/CC-BY-4.0.html",
    "dateModified": "2026-08-08",
    "citation": [
      {
        "@type": "CreativeWork",
        "name": "agent-reliability editorial brief and blueprint (Gate 1 approved, 2026-08-08)"
      },
      {
        "@type": "CreativeWork",
        "name": "OWASP Top 10 for LLM Applications and Generative AI (2025)",
        "url": "https://genai.owasp.org/llm-top-10/"
      },
      {
        "@type": "CreativeWork",
        "name": "Building effective agents",
        "url": "https://www.anthropic.com/research/building-effective-agents"
      }
    ],
    "author": {
      "@type": "Person",
      "name": "Santiago Santa María Morales",
      "jobTitle": "practitioner — harness engineering and agent evaluation in production"
    },
    "@type": "Article",
    "headline": "Human approval gates and graduated autonomy"
  }
}