{
  "id": "behavioral-canaries",
  "type": "entity",
  "name": "Behavioral canaries",
  "summary": "Small, fixed probe tasks run continuously against a deployed agent. Each canary has a known-good outcome; a change in that outcome is treated as an early regression signal — model drift, prompt rot, tool API changes — before users feel it. The chaos-engineering steady-state hypothesis applied to agent behavior instead of infrastructure metrics.",
  "locale": "en",
  "tags": [
    "monitoring",
    "canaries",
    "regression",
    "production"
  ],
  "relations": [
    {
      "rel": "related",
      "target": "guardian-falsification"
    },
    {
      "rel": "related",
      "target": "agent-telemetry-actor-classification"
    }
  ],
  "questions": [
    "How do I notice my agent regressed before my users do?",
    "What is a behavioral canary?"
  ],
  "claims": [
    {
      "id": "c1",
      "text": "A behavioral canary is a probe task with a pinned expected outcome, executed on a schedule against the production agent; outcome drift is an alarm, and the practice is part of the production methodology behind this instance.",
      "sources": [
        "citarium-methodology-brief"
      ],
      "confidence": 0.85
    },
    {
      "id": "c2",
      "text": "Chaos engineering grounds the pattern: define steady state as measurable output that indicates normal behavior, then treat deviations from that baseline as the signal worth alarming on.",
      "sources": [
        "principles-chaos"
      ],
      "confidence": 0.9
    }
  ],
  "takeaways": [],
  "faqs": [],
  "evidence_tier": "secondary",
  "evidence": {
    "level": "production",
    "source_types": [
      "production_system",
      "personal_experience",
      "industry_observation"
    ]
  },
  "moat_flag": true,
  "winning_edge": "Transfers a proven infrastructure discipline (steady-state probing) to agent behavior with the operational details — pinned outcomes, schedule, drift-as-alarm — from a team that runs it, not from analogy alone.",
  "confidence": 0.8,
  "last_verified": "2026-08-08",
  "canonical_url": "https://agentreliability.dev/k/behavioral-canaries",
  "api_url": "https://agentreliability.dev/api/k/behavioral-canaries.json",
  "jsonld": {
    "@context": "https://schema.org",
    "name": "Behavioral canaries",
    "description": "Small, fixed probe tasks run continuously against a deployed agent. Each canary has a known-good outcome; a change in that outcome is treated as an early regression signal — model drift, prompt rot, tool API changes — before users feel it. The chaos-engineering steady-state hypothesis applied to agent behavior instead of infrastructure metrics.",
    "url": "https://agentreliability.dev/k/behavioral-canaries",
    "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": "Principles of Chaos Engineering",
        "url": "https://principlesofchaos.org/"
      }
    ],
    "author": {
      "@type": "Person",
      "name": "Santiago Santa María Morales",
      "jobTitle": "practitioner — harness engineering and agent evaluation in production"
    },
    "@type": "Article",
    "headline": "Behavioral canaries"
  }
}