{
  "schema_version": "1.1",
  "id": "archive:https://www.youtube.com/watch?v=shRR1e2HXMk",
  "slug": "codex-behind-the-harness-dominik-kundel-openai-04cntno",
  "url": "https://feed7.dev/p/codex-behind-the-harness-dominik-kundel-openai-04cntno",
  "title": "Codex, Behind the Harness — Dominik Kundel, OpenAI",
  "why_included": "Codex exposes several reusable harness patterns: deferred tools, persistent browser execution, sandboxed edits, automated permission review, WebSockets, and server-side compaction.",
  "summary": "Codex’s open-source Rust harness caps skill descriptions at **2% of the context window** and defers some tools until search finds them. Since **GPT-5.4**, deferred tool loading is also available through the Responses API.",
  "practical_implication": "For custom coding agents, borrow the boundaries rather than rebuilding every mechanism: persistent execution for browser scripts, apply-patch for edits, OS-specific sandboxes, read-only review agents for escalations, and server-side compaction for long runs.",
  "agent_context": "Codex’s open-source Rust harness caps skill descriptions at **2% of the context window** and defers some tools until search finds them. Since **GPT-5.4**, deferred tool loading is also available through the Responses API.\n\nFor custom coding agents, borrow the boundaries rather than rebuilding every mechanism: persistent execution for browser scripts, apply-patch for edits, OS-specific sandboxes, read-only review agents for escalations, and server-side compaction for long runs.\n\nAt **1,000 tokens per second**, GPT-5.3 Codex Spark made network round trips the bottleneck, motivating a persistent WebSocket mode. These details describe the current harness and API surface; both are expected to shift with later models.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=shRR1e2HXMk",
    "published_at": "2026-08-10T16:00:06.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "agent",
  "domains": [
    "coding",
    "security"
  ],
  "topics": [
    "harness-engineering",
    "context-engineering",
    "sandboxing"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "At **1,000 tokens per second**, GPT-5.3 Codex Spark made network round trips the bottleneck, motivating a persistent WebSocket mode. These details describe the current harness and API surface; both are expected to shift with later models."
  ],
  "connected_context": {
    "meaning": "This turns several broad harness recommendations into concrete Codex boundaries: budget skill discovery, defer tools, preserve execution, constrain edits with sandboxes and review, and compact server-side. It reinforces progressive loading, isolation, and long-run context controls in the candidates, while adding a latency consequence: once inference is extremely fast, connection architecture becomes part of agent performance. The mechanisms remain model- and API-version dependent.",
    "corpus_size": 419,
    "generated_at": "2026-08-12T10:05:12.509Z",
    "connections": [
      {
        "title": "Skills are new features: Building Skill-Centric Harness — Yogendra Miraje, FactSet",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=7jjudsEhBtM",
        "feed7_url": "https://feed7.dev/p/skills-are-new-features-building-skill-centric-harness-yogendra-miraje-f-0lp4c7o",
        "reason": "FactSet frames progressive skill loading and sandboxed execution as governance requirements; Codex provides concrete implementations through a context budget, deferred discovery, OS sandboxes, and review escalation."
      },
      {
        "title": "Evolution of agentic surfaces — Gagan Bhat & Isabella Kai He, Anthropic",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=K0X9QDRkIdg",
        "feed7_url": "https://feed7.dev/p/evolution-of-agentic-surfaces-gagan-bhat-isabella-kai-he-anthropic-0mtshqh",
        "reason": "Both show that harness behavior is coupled to model releases; Anthropic’s obsolete reset workaround is direct evidence for retesting Codex’s current compaction and loading choices as models change."
      },
      {
        "title": "Anthropic's CCA Exam as a Field-Guide for Agentic Engineering — Frank Coyle, UC Berkeley",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=Z-c11pV_uvU",
        "feed7_url": "https://feed7.dev/p/anthropic-s-cca-exam-as-a-field-guide-for-agentic-engineering-frank-coyl-00nf896",
        "reason": "The candidate’s compaction, termination inspection, and constrained-agent guidance is reinforced by Codex’s server-side compaction and read-only review boundary."
      },
      {
        "title": "Multiplayer agentic engineering — Arjun Singh, Superconductor",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=OL7kfezynJM",
        "feed7_url": "https://feed7.dev/p/multiplayer-agentic-engineering-arjun-singh-superconductor-01w4k91",
        "reason": "Superconductor’s isolated cloud sessions establish isolation as a team workflow requirement; Codex adds lower-level sandbox and permission-review mechanisms for enforcing that boundary."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-08-10T16:00:06.000Z",
  "modified_at": "2026-08-10T16:00:06.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/codex-behind-the-harness-dominik-kundel-openai-04cntno",
    "json": "https://feed7.dev/p/codex-behind-the-harness-dominik-kundel-openai-04cntno.json",
    "markdown": "https://feed7.dev/p/codex-behind-the-harness-dominik-kundel-openai-04cntno.md"
  }
}