Update to Repo Scout

{
  "canonical_app_id": null,
  "connector_id": null,
  "created_at": "2026-09-08T19:15:33.816401Z",
  "discoverability": "LISTED",
  "id": "plugins_6aa0595d1d1881918d6321ebc0d66b91",
  "is_template": false,
  "name": "repo-scout",
  "release": {
    "app_ids": [],
    "app_manifest": null,
    "app_templates": [],
    "description": "Evidence-backed engineering backlog discovery: turn any repository into a prioritized list of real work without manufacturing issues.",
    "display_name": "Repo Scout",
    "id": "pluginrel_cf6e2682c3a481918e133ff7b2025f87",
    "interface": {
      "brand_color": "#7C3AED",
      "capabilities": [
        "Interactive",
        "Read"
      ],
      "category": "Productivity",
      "composer_icon_dark_url": null,
      "composer_icon_url": "https://files.openai.com/content?id=file_000000002788820997348989d7ab9e42",
      "default_prompt": "Find evidence-backed engineering work in this repo.",
      "default_prompts": [
        "Find evidence-backed engineering work in this repo.",
        "Scan this repository; separate proven bugs from unverified risks.",
        "Draft a prioritized backlog without changing code or opening issues."
      ],
      "developer_name": "YU-SHANG LUNG",
      "logo_url": "https://files.openai.com/content?id=file_00000000a944820ea2c7120c265f88a1",
      "logo_url_dark": null,
      "long_description": "Repo Scout reviews a whole repository, not just a diff: architecture, frontend, backend and data, mobile, systems, security and privacy, tests and CI, docs and releases, and AI-agent code. It maps modules and data flow first, then hunts for candidate problems, actively looks for counter-evidence, and returns a ranked backlog where every finding carries a source location, impact, an evidence level (E0 lead to E3 reproduced), the alternative explanations that were checked, a minimal fix direction and acceptance criteria. The report ends with a coverage matrix that says what was inspected, partially inspected, blocked or skipped. The default scan mode never modifies code, runs project scripts, opens issues or triggers CI; verification, planning, publishing and fixing each need explicit authorization. A read-only, filename-only inventory helper is bundled; no MCP server, credentials, scheduler or background service is included.",
      "plugin_category_id": "productivity",
      "privacy_policy_url": "https://iml1s.github.io/repo-scout/privacy/",
      "screenshot_urls": [],
      "short_description": "Find real work in any repo",
      "terms_of_service_url": "https://iml1s.github.io/repo-scout/terms/",
      "website_url": "https://iml1s.github.io/repo-scout/"
    },
    "keywords": [
      "repo scout",
      "repo-scout",
      "engineering backlog",
      "code audit",
      "evidence-backed review"
    ],
    "onboarding_skill_name": null,
    "requires_local_executor": false,
    "skills": [
      {
        "description": "Find evidence-backed engineering work in an existing repository across languages, architecture, frontend, backend, mobile, systems, data, security, delivery and AI agents. Use for whole-repo audits, maintenance backlogs, verified improvement plans, or requests to find useful work in source code. Do not manufacture issues, assume universal analyzer support, or begin repairs without authorization.",
        "interface": {
          "brand_color": null,
          "default_prompt": "Use $repo-scout to inspect this repository and draft evidence-backed improvements without modifying code or publishing issues.",
          "display_name": "Repo Scout — 裝忙 Skills",
          "icon_large_url": null,
          "icon_small_url": null,
          "iconography": "radar",
          "short_description": "Find evidence-backed engineering work across a repository"
        },
        "name": "repo-scout",
        "plugin_release_skill_id": "pluginrsk_6aa16bb9cc508191b34d5d6b024a86d5"
      }
    ],
    "version": "0.1.2"
  },
  "scope": "GLOBAL",
  "status": "ENABLED"
}