{
  "schemaVersion": "1.0.0",
  "id": "https://aisdlc.ai/insights/catalog.json",
  "title": "AISDLC Insights editorial catalog",
  "edition": "2026.08",
  "reviewedAt": "2026-08-09",
  "authorityBoundary": "AISDLC Insights publishes source-informed editorial synthesis and implementation positions. It is reference material, not a standard, certification, legal opinion, or authorization to deploy an agent.",
  "author": {
    "id": "aisdlc-editorial",
    "name": "AISDLC Editorial",
    "kind": "Organization",
    "description": "The organizational editorial function for the Agentic Engineering Atlas and the Agentic AI SDLC reference framework."
  },
  "articles": [
    {
      "id": "the-model-is-not-the-architecture",
      "canonicalUrl": "https://aisdlc.ai/insights/the-model-is-not-the-architecture",
      "index": "I01",
      "title": "The model is not the architecture",
      "deck": "Enterprise advantage will not come from renting the same frontier model as everyone else. It will come from engineering the system that turns capability into bounded, repeatable, provable work.",
      "thesis": "Capability is rented. Intent, context, identity, tools, verification, evidence, and decision rights are engineered.",
      "category": "Operating thesis",
      "audiences": [
        "Executive",
        "Engineering",
        "Platform"
      ],
      "readingMinutes": 7,
      "publishedAt": "2026-08-08",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "openai-harness-engineering",
        "anthropic-managed-agents",
        "google-dora-ai-development",
        "karpathy-software-3"
      ],
      "conceptIds": [
        "agentic-engineering",
        "harness-engineering",
        "execution-substrate",
        "evidence-engineering"
      ]
    },
    {
      "id": "context-is-a-runtime-system",
      "canonicalUrl": "https://aisdlc.ai/insights/context-is-a-runtime-system",
      "index": "I02",
      "title": "Context is a runtime system",
      "deck": "Prompt craft still matters. Production reliability now depends on the full state made visible, trusted, current, and actionable at each decision.",
      "thesis": "The control surface is everything the agent can see—and every rule that governs how it got there.",
      "category": "Context systems",
      "audiences": [
        "Engineering",
        "Platform",
        "Risk"
      ],
      "readingMinutes": 7,
      "publishedAt": "2026-08-08",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "anthropic-context",
        "anthropic-long-running-harness",
        "owasp-memory",
        "github-spec-persistence"
      ],
      "conceptIds": [
        "context-engineering",
        "durable-memory",
        "intent-engineering"
      ]
    },
    {
      "id": "harness-engineering-is-platform-engineering",
      "canonicalUrl": "https://aisdlc.ai/insights/harness-engineering-is-platform-engineering",
      "index": "I03",
      "title": "Harness engineering is platform engineering",
      "deck": "The model supplies capability. The harness determines whether that capability can plan, act, recover, prove, and stop inside a real engineering environment.",
      "thesis": "The harness is the operating system for bounded agent work.",
      "category": "Platform architecture",
      "audiences": [
        "Engineering",
        "Platform",
        "Operations"
      ],
      "readingMinutes": 8,
      "publishedAt": "2026-08-08",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "openai-harness-engineering",
        "anthropic-long-running-harness",
        "anthropic-harness-design",
        "google-agent-substrate",
        "openai-symphony"
      ],
      "conceptIds": [
        "harness-engineering",
        "execution-substrate",
        "tools-and-protocols",
        "durable-memory"
      ]
    },
    {
      "id": "verification-is-a-plane-not-a-phase",
      "canonicalUrl": "https://aisdlc.ai/insights/verification-is-a-plane-not-a-phase",
      "index": "I04",
      "title": "Verification is a plane, not a phase",
      "deck": "A builder can test its own work. It cannot independently accept the risk created by that work.",
      "thesis": "Independent challenge must cross the lifecycle, not wait at the end of it.",
      "category": "Independent assurance",
      "audiences": [
        "Risk",
        "Engineering",
        "Executive"
      ],
      "readingMinutes": 7,
      "publishedAt": "2026-08-08",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "anthropic-evals",
        "anthropic-harness-design",
        "nist-benchmark-evals",
        "anthropic-infrastructure-noise",
        "openai-monitoring"
      ],
      "conceptIds": [
        "multi-agent-verification",
        "eval-driven-development",
        "observability-control"
      ]
    },
    {
      "id": "every-agent-is-a-governed-principal",
      "canonicalUrl": "https://aisdlc.ai/insights/every-agent-is-a-governed-principal",
      "index": "I05",
      "title": "Every agent is a governed principal",
      "deck": "If an agent can take consequential action, it needs its own attributable identity, a delegator, a purpose, an authority envelope, and an expiry.",
      "thesis": "An agent is neither a person nor a conventional application. Its identity lifecycle must expose who is accountable for every delegated action.",
      "category": "Identity and authority",
      "audiences": [
        "Risk",
        "Platform",
        "Operations",
        "Executive"
      ],
      "readingMinutes": 7,
      "publishedAt": "2026-08-08",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "nist-agent-identity",
        "nist-agent-standards",
        "microsoft-agent-identities",
        "mcp-enterprise-auth",
        "owasp-agentic-top10"
      ],
      "conceptIds": [
        "agent-identity",
        "tools-and-protocols",
        "observability-control",
        "execution-substrate"
      ]
    },
    {
      "id": "evidence-before-autonomy",
      "canonicalUrl": "https://aisdlc.ai/insights/evidence-before-autonomy",
      "index": "I06",
      "title": "Evidence before autonomy",
      "deck": "Autonomy should expand only when the organization can prove what the system did, why it was allowed, how it was challenged, and how it can be stopped.",
      "thesis": "Autonomy is an evidence-backed operating state, not a permanent product feature.",
      "category": "Evidence and control",
      "audiences": [
        "Executive",
        "Risk",
        "Operations",
        "Engineering"
      ],
      "readingMinutes": 8,
      "publishedAt": "2026-08-08",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "slsa-provenance",
        "openai-self-improving",
        "nist-deployed-monitoring",
        "opentelemetry-genai",
        "openai-monitoring"
      ],
      "conceptIds": [
        "evidence-engineering",
        "observability-control",
        "eval-driven-development",
        "agent-identity"
      ]
    },
    {
      "id": "own-the-speed",
      "canonicalUrl": "https://aisdlc.ai/insights/own-the-speed",
      "index": "I07",
      "title": "Own the speed",
      "deck": "The goal is not to slow generation. It is to redesign the system that turns generated work into trusted change.",
      "thesis": "When implementation accelerates, the organization must expand its capacity to decide, verify, integrate, release, and learn.",
      "category": "Operating thesis",
      "audiences": [
        "Executive",
        "Engineering",
        "Operations",
        "Risk"
      ],
      "readingMinutes": 8,
      "publishedAt": "2026-08-09",
      "reviewedAt": "2026-08-09",
      "sourceIds": [
        "ai-engineer-velocity-sickness",
        "google-dora-ai-development",
        "dora-wip-limits",
        "dora-small-batches",
        "space-developer-productivity",
        "github-agent-pr-review",
        "metr-developer-productivity-rct"
      ],
      "conceptIds": [
        "agentic-engineering",
        "intent-engineering",
        "harness-engineering",
        "multi-agent-verification",
        "human-accountability"
      ]
    },
    {
      "id": "train-the-judgment-not-the-tool",
      "canonicalUrl": "https://aisdlc.ai/insights/train-the-judgment-not-the-tool",
      "index": "I08",
      "title": "Train the judgment, not the tool",
      "deck": "Models and interfaces will keep changing. The durable capability is knowing what to delegate, how to bound it, what evidence to demand, and when a human must say no.",
      "thesis": "Agentic-engineering education should make judgment visible in independently challengeable work—not equate content completion with competence or authority.",
      "category": "Operating thesis",
      "audiences": [
        "Executive",
        "Engineering",
        "Risk"
      ],
      "readingMinutes": 7,
      "publishedAt": "2026-08-11",
      "reviewedAt": "2026-08-11",
      "sourceIds": [
        "metr-developer-productivity-rct",
        "google-dora-ai-development",
        "space-developer-productivity",
        "karpathy-software-3"
      ],
      "conceptIds": [
        "agentic-engineering",
        "multi-agent-verification",
        "evidence-engineering",
        "human-accountability"
      ]
    }
  ]
}