{
  "schema_version": "newruntime-agent-readable-v0.1",
  "type": "raw_signal",
  "id": "tg-2409",
  "slug": "agent-evals-become-a-discipline-separate-from-model-evals",
  "title": "Agent Evals Become a Discipline Separate from Model Evals",
  "description": "Reliable agents require evaluation of trajectories, tools, state, recovery, and outcomes rather than grading only the final answer.",
  "observed_at": "2026-05-21",
  "why_it_matters": "This dated record adds public evidence to the verification bandwidth is the scarce resource, skills become portable capability layer analysis and keeps the claim auditable as the underlying products and practices change.",
  "novelty": "notable",
  "verification_level": "source-linked",
  "signal_type": "research",
  "evidence_kind": "independent-analysis",
  "status": "published",
  "telegram_message_id": 2409,
  "telegram_url": "https://t.me/qwgai/2409",
  "topics": [
    "agent-evals",
    "reliability",
    "tool-use"
  ],
  "entities": [],
  "related_patterns": [
    "verification-bandwidth-is-the-scarce-resource",
    "skills-become-portable-capability-layer"
  ],
  "source_urls": [
    "https://cameronrwolfe.substack.com/p/agent-evals"
  ],
  "import_batch": "telegram-2026-07-17-plus-2026-07-20-v2",
  "routes": {
    "html": "https://newruntime.com/signals/agent-evals-become-a-discipline-separate-from-model-evals/",
    "markdown": "https://newruntime.com/signals/agent-evals-become-a-discipline-separate-from-model-evals.md",
    "json": "https://newruntime.com/signals/agent-evals-become-a-discipline-separate-from-model-evals.json"
  },
  "source_format": "telegram-export-normalized-json"
}