Update to Codex Advisor

{
  "canonical_app_id": null,
  "connector_id": null,
  "created_at": "2026-09-02T16:40:23.692958Z",
  "discoverability": "LISTED",
  "id": "plugins_6a984f37e9c88191a2a777998f7b0521",
  "is_template": false,
  "is_template_instance": false,
  "is_v2_template": false,
  "name": "advisor",
  "release": {
    "app_ids": [],
    "app_manifest": null,
    "app_templates": [],
    "description": "One fresh, read-only consultation for material technical decisions in Codex CLI and Codex desktop.",
    "display_name": "Codex Advisor",
    "id": "pluginrel_9f3d2106268481919a0b3f2eda1bb5dd",
    "interface": {
      "brand_color": null,
      "capabilities": [
        "Interactive",
        "Read"
      ],
      "category": "Developer Tools",
      "composer_icon_dark_url": null,
      "composer_icon_url": "https://files.openai.com/content?id=file_00000000260481fbb7d9a60aec0e4fab",
      "default_prompt": "Should we split this service into microservices?",
      "default_prompts": [
        "Should we split this service into microservices?",
        "Review this database migration plan.",
        "Which API design is safer for this integration?"
      ],
      "developer_name": "David Schmidt / Zero Delta LLC",
      "logo_url": "https://files.openai.com/content?id=file_000000003c8881f899457fdbac90fcbe",
      "logo_url_dark": null,
      "long_description": "Codex Advisor helps software developers review architecture, API interfaces, data models, and migration plans in Codex CLI and Codex desktop. Each consultation returns one recommendation with tradeoffs, risks, and acceptance checks.\n\nAdvisor sends bounded technical context to OpenAI for inference through your authenticated account and allowance; Zero Delta runs no relay or telemetry backend and does not automatically receive consultation packets or responses. Redacted receipts you voluntarily share through support channels are covered by the privacy policy. The consultation child is launched in a read-only sandbox and instructed to use zero tools; after execution the launcher inspects the recorded runtime and rejects the result if it observes any tool call, which is a post-execution acceptance check, not preventive tool-free isolation. The local launcher writes temporary files and Codex session records. The root agent owns implementation and final decisions; Advisor does not implement, deploy, or perform final review.\n\nRequires Python 3.11+, a local Codex host, jq, a supported POSIX shell, and a persisted session. Standard uses Terra/high, Specialist uses GPT-6.1 Sol/high, and explicit-only Astra/high is available for the most complex uses.",
      "plugin_category_id": "developer tools",
      "privacy_policy_url": "https://zerodelta.dev/advisor/privacy/",
      "screenshot_urls": [],
      "short_description": "Architecture and code advice.",
      "terms_of_service_url": "https://zerodelta.dev/advisor/terms/",
      "website_url": "https://zerodelta.dev/advisor/"
    },
    "keywords": [
      "codex",
      "advisor",
      "architecture-review",
      "second-opinion",
      "decision-support",
      "read-only"
    ],
    "mcp_servers": [],
    "onboarding_skill_name": null,
    "requires_local_executor": false,
    "skills": [
      {
        "description": "Consult for ordinary material architecture, interface, data-model, or explicit advisor, generic-advisor, authorization, challenge, second-opinion, or architecture-review choices; or, when targeted evidence still leaves unresolved cross-module or system design, compatibility or concurrency boundaries, competing diagnoses, security, privacy, or trust boundaries, recovery, irreversible-state migration, or data-loss decisions; and the completion consultation that complex multi-phase or multi-file work takes before it is declared complete. Skip factual/status/summarization work, fully determined mechanical edits, formatting/renaming/docs synchronization, settled-plan execution, final review owned elsewhere, explicit no-delegation/root-only requests, and every borderline case.",
        "interface": {
          "brand_color": null,
          "default_prompt": "Use $consultation for one bounded technical decision, then record whether the advice is accepted, modified, or rejected.",
          "display_name": "Advisor Consultation",
          "icon_large_url": null,
          "icon_small_url": null,
          "iconography": "hierarchy",
          "short_description": "Fresh read-only advice for material technical decisions"
        },
        "name": "consultation",
        "plugin_release_skill_id": "pluginrsk_6ae1528647888191908c986622e0808b"
      }
    ],
    "version": "1.4.10"
  },
  "scope": "GLOBAL",
  "share_url": "https://chatgpt.com/plugins/plugins_6a984f37e9c88191a2a777998f7b0521?open_in_app",
  "status": "ENABLED"
}