[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78DF6CKZ6Y19XWFFAFRMPVA":3,"skill-same:01M78DF6CKZ6Y19XWFFAFRMPVA":43},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":28,"install":29,"resources":42},"01M78DF6CKZ6Y19XWFFAFRMPVA","drive-openhub",null,"Use to connect to and drive a OpenHub instance from an external agent (Claude Code \u002F Cursor \u002F any MCP client). Delegate goals to OpenHub's autonomous agent, drive its browser \u002F computer \u002F knowledge base \u002F files, and manage the platform — over MCP or REST, with a per-companion access token (you run AS the bound companion). Use this whenever the user asks to control, automate, or hand work off to \"OpenHub\", \"their OpenHub\", \"the desktop companion\", or a running OpenHub server.","Grong\u002Fopenhub-client","Grong","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F4919858?v=4","01M78DABC3WYM9Z67DC6JSDGQF","01M78DABD6RAC4TNESYHD8WT4B",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fdrive-nomifun\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: drive-openhub\ndescription: >-\n  Use to connect to and drive a OpenHub instance from an external agent\n  (Claude Code \u002F Cursor \u002F any MCP client). Delegate goals to OpenHub's\n  autonomous agent, drive its browser \u002F computer \u002F knowledge base \u002F files, and\n  manage the platform — over MCP or REST, with a per-companion access token (you\n  run AS the bound companion). Use this whenever the user asks to control,\n  automate, or hand work off to \"OpenHub\", \"their OpenHub\", \"the desktop\n  companion\", or a running OpenHub server.\n---\n\n# Drive OpenHub (external companion)\n\nOpenHub exposes its full platform capability set — an autonomous agent, browser\nautomation, computer control, knowledge bases, files, terminals, and platform\nmanagement — to external callers through one **MCP** endpoint and an equivalent\n**REST** API, authenticated by a **per-companion access token**. Each token is\nbound to one specific companion: calling with it runs you **as that companion**,\ninheriting its profile model \u002F persona \u002F knowledge bases. Connecting makes you\nan \"external companion\": you drive the platform exactly as its built-in desktop\ncompanion does.\n\n## 1. Connect (MCP, recommended)\n\nConfigure OpenHub as a Streamable-HTTP MCP server:\n\n```json\n{\n  \"mcpServers\": {\n    \"openhub\": {\n      \"type\": \"streamable-http\",\n      \"url\": \"http:\u002F\u002F\u003Chost>:25808\u002Fmcp-agent\",\n      \"headers\": { \"Authorization\": \"Bearer \u003Ctoken>\" }\n    }\n  }\n}\n```\n\n- `\u003Chost>` is the machine running OpenHub (`127.0.0.1` locally, or its LAN\u002Fpublic\n  address with WebUI remote access enabled).\n- **`\u002Fmcp-agent`** advertises a tight, curated tool set for getting work done\n  (agent delegation, browser, computer, knowledge, files). Use **`\u002Fmcp`** instead\n  for the full platform-control surface (~140 tools incl. channels, companions,\n  cron, providers, …).\n- `\u003Ctoken>` is a **per-companion access token** — it binds you to one companion.\n  Get it from the OpenHub operator: in the desktop WebUI\u002Fremote panel, or by\n  minting one (see \"Minting a token\" below). The companion you bind to **must\n  have a configured model** for `openhub_agent_run` (and other model-backed caps) to\n  work — otherwise minting returns a `warning` and those calls fail until a model\n  is set in Model Management.\n\nREST equivalent (for scripts): `POST http:\u002F\u002F\u003Chost>:25808\u002Fv1\u002Ftools\u002F\u003Cname>` with the\nsame Bearer token; `GET \u002Fv1\u002Ftools?profile=agent` lists tools; `GET\n\u002Fv1\u002Fopenapi.json?profile=agent` is a machine-readable contract.\n\n### Minting a token (operator, local-trust only)\n\nThe mint\u002Fquery\u002Frevoke endpoints are **local-trust gated** (reachable only from\nthe desktop client \u002F loopback, not from a remote browser). `{id}` is the\ncompanion id you want the token bound to:\n\n```bash\n# Mint — returns the plaintext token ONCE, bound to that companion.\ncurl -X POST http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fapi\u002Fwebui\u002Fcompanions\u002F\u003Cid>\u002Faccess-token\n# => { \"token\": \"\u003C64-hex token>\", \"companion_id\": \"\u003Cid>\" }\n#    If that companion has no usable model yet, the body also carries\n#    \"warning\": \"…\" (the token is still minted, but openhub_agent_run etc. will fail\n#    until you configure a model).\n\n# Query whether one is configured (does NOT return the token):\ncurl http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fapi\u002Fwebui\u002Fcompanions\u002F\u003Cid>\u002Faccess-token\n# => { \"configured\": true }\n\n# Revoke:\ncurl -X DELETE http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fapi\u002Fwebui\u002Fcompanions\u002F\u003Cid>\u002Faccess-token\n```\n\nFor a **headless** server, seed a token at startup via the\n`OPENHUB_COMPANION_TOKEN` env var — it binds to the **default companion** (only\nif no token is configured yet; it won't overwrite an existing one):\n\n```bash\nOPENHUB_COMPANION_TOKEN=\"$(openssl rand -hex 32)\" openhub-core   # or openhub-web\n```\n\n## 2. Delegate a goal (the headline move)\n\nTo hand a whole task to OpenHub's own autonomous agent, call **`openhub_agent_run`**:\n\n```json\n{ \"goal\": \"Research X and write a summary to notes.md\", \"timeout_secs\": 600 }\n```\n\nIt spins up a fresh autonomous OpenHub agent (full tools, using the **bound\ncompanion's** profile model — which must be configured), runs the goal to\ncompletion, and returns `{ \"status\": \"completed\", \"text\": \"\u003Cfinal answer>\", \"conversation_id\": \u003Cid> }`.\nFor long tasks it returns `{ \"status\": \"running\", \"conversation_id\": \u003Cid> }` —\npoll **`openhub_agent_result`** with that `conversation_id` until completed.\n\n## 3. Or drive capabilities directly\n\nUse individual tools when you want fine control: `openhub_browser_*` (navigate \u002F\nobserve \u002F act), the computer tools, `openhub_knowledge_*` (search \u002F read \u002F write\nknowledge bases), `openhub_fs_*` (read \u002F write \u002F browse files), `openhub_create_terminal`,\nand the conversation tools. `GET \u002Fv1\u002Ftools` (or MCP `tools\u002Flist`) is the live,\nauthoritative catalog with JSON Schemas.\n\n## 4. Confirmations & limits\n\n- **Destructive actions** (deletes, etc.) return `{ \"needs_confirmation\": true,\n  \"restate\": \"...\" }`. Restate the exact action to the user, get agreement, then\n  re-call the same tool with `\"confirm\": true`.\n- **Sensitive actions** (secrets, factory reset) are **denied** on this surface.\n- **Trust model:** holding a per-companion access token grants full,\n  RCE-equivalent control of that OpenHub instance (as the bound companion). Treat\n  the token as a high-value secret; only connect to instances you are authorized\n  to drive. Revoking a token affects only its companion; other companions' tokens\n  are unaffected.\n\n## 5. Failure handling\n\n- `401` → missing \u002F invalid \u002F revoked access token.\n- REST `409` (or `needs_confirmation` in the body) → re-call with `confirm: true`.\n- REST `422` \u002F a `{ \"error\": ... }` body → the tool rejected the arguments;\n  check the schema from `\u002Fv1\u002Ftools` and retry.\n- Connection refused → OpenHub isn't running or the URL\u002Fport is wrong.\n","# Drive OpenHub (external companion)\n\nOpenHub exposes its full platform capability set — an autonomous agent, browser\nautomation, computer control, knowledge bases, files, terminals, and platform\nmanagement — to external callers through one **MCP** endpoint and an equivalent\n**REST** API, authenticated by a **per-companion access token**. Each token is\nbound to one specific companion: calling with it runs you **as that companion**,\ninheriting its profile model \u002F persona \u002F knowledge bases. Connecting makes you\nan \"external companion\": you drive the platform exactly as its built-in desktop\ncompanion does.\n\n## 1. Connect (MCP, recommended)\n\nConfigure OpenHub as a Streamable-HTTP MCP server:\n\n```json\n{\n  \"mcpServers\": {\n    \"openhub\": {\n      \"type\": \"streamable-http\",\n      \"url\": \"http:\u002F\u002F\u003Chost>:25808\u002Fmcp-agent\",\n      \"headers\": { \"Authorization\": \"Bearer \u003Ctoken>\" }\n    }\n  }\n}\n```\n\n- `\u003Chost>` is the machine running OpenHub (`127.0.0.1` locally, or its LAN\u002Fpublic\n  address with WebUI remote access enabled).\n- **`\u002Fmcp-agent`** advertises a tight, curated tool set for getting work done\n  (agent delegation, browser, computer, knowledge, files). Use **`\u002Fmcp`** instead\n  for the full platform-control surface (~140 tools incl. channels, companions,\n  cron, providers, …).\n- `\u003Ctoken>` is a **per-companion access token** — it binds you to one companion.\n  Get it from the OpenHub operator: in the desktop WebUI\u002Fremote panel, or by\n  minting one (see \"Minting a token\" below). The companion you bind to **must\n  have a configured model** for `openhub_agent_run` (and other model-backed caps) to\n  work — otherwise minting returns a `warning` and those calls fail until a model\n  is set in Model Management.\n\nREST equivalent (for scripts): `POST http:\u002F\u002F\u003Chost>:25808\u002Fv1\u002Ftools\u002F\u003Cname>` with the\nsame Bearer token; `GET \u002Fv1\u002Ftools?profile=agent` lists tools; `GET\n\u002Fv1\u002Fopenapi.json?profile=agent` is a machine-readable contract.\n\n### Minting a token (operator, local-trust only)\n\nThe mint\u002Fquery\u002Frevoke endpoints are **local-trust gated** (reachable only from\nthe desktop client \u002F loopback, not from a remote browser). `{id}` is the\ncompanion id you want the token bound to:\n\n```bash\n# Mint — returns the plaintext token ONCE, bound to that companion.\ncurl -X POST http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fapi\u002Fwebui\u002Fcompanions\u002F\u003Cid>\u002Faccess-token\n# => { \"token\": \"\u003C64-hex token>\", \"companion_id\": \"\u003Cid>\" }\n#    If that companion has no usable model yet, the body also carries\n#    \"warning\": \"…\" (the token is still minted, but openhub_agent_run etc. will fail\n#    until you configure a model).\n\n# Query whether one is configured (does NOT return the token):\ncurl http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fapi\u002Fwebui\u002Fcompanions\u002F\u003Cid>\u002Faccess-token\n# => { \"configured\": true }\n\n# Revoke:\ncurl -X DELETE http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fapi\u002Fwebui\u002Fcompanions\u002F\u003Cid>\u002Faccess-token\n```\n\nFor a **headless** server, seed a token at startup via the\n`OPENHUB_COMPANION_TOKEN` env var — it binds to the **default companion** (only\nif no token is configured yet; it won't overwrite an existing one):\n\n```bash\nOPENHUB_COMPANION_TOKEN=\"$(openssl rand -hex 32)\" openhub-core   # or openhub-web\n```\n\n## 2. Delegate a goal (the headline move)\n\nTo hand a whole task to OpenHub's own autonomous agent, call **`openhub_agent_run`**:\n\n```json\n{ \"goal\": \"Research X and write a summary to notes.md\", \"timeout_secs\": 600 }\n```\n\nIt spins up a fresh autonomous OpenHub agent (full tools, using the **bound\ncompanion's** profile model — which must be configured), runs the goal to\ncompletion, and returns `{ \"status\": \"completed\", \"text\": \"\u003Cfinal answer>\", \"conversation_id\": \u003Cid> }`.\nFor long tasks it returns `{ \"status\": \"running\", \"conversation_id\": \u003Cid> }` —\npoll **`openhub_agent_result`** with that `conversation_id` until completed.\n\n## 3. Or drive capabilities directly\n\nUse individual tools when you want fine control: `openhub_browser_*` (navigate \u002F\nobserve \u002F act), the computer tools, `openhub_knowledge_*` (search \u002F read \u002F write\nknowledge bases), `openhub_fs_*` (read \u002F write \u002F browse files), `openhub_create_terminal`,\nand the conversation tools. `GET \u002Fv1\u002Ftools` (or MCP `tools\u002Flist`) is the live,\nauthoritative catalog with JSON Schemas.\n\n## 4. Confirmations & limits\n\n- **Destructive actions** (deletes, etc.) return `{ \"needs_confirmation\": true,\n  \"restate\": \"...\" }`. Restate the exact action to the user, get agreement, then\n  re-call the same tool with `\"confirm\": true`.\n- **Sensitive actions** (secrets, factory reset) are **denied** on this surface.\n- **Trust model:** holding a per-companion access token grants full,\n  RCE-equivalent control of that OpenHub instance (as the bound companion). Treat\n  the token as a high-value secret; only connect to instances you are authorized\n  to drive. Revoking a token affects only its companion; other companions' tokens\n  are unaffected.\n\n## 5. Failure handling\n\n- `401` → missing \u002F invalid \u002F revoked access token.\n- REST `409` (or `needs_confirmation` in the body) → re-call with `confirm: true`.\n- REST `422` \u002F a `{ \"error\": ... }` body → the tool rejected the arguments;\n  check the schema from `\u002Fv1\u002Ftools` and retry.\n- Connection refused → OpenHub isn't running or the URL\u002Fport is wrong.\n",{"name":5,"description":7},"main","89ebf7196eee1581caf96eeee57dd956e5b7c6f2","docs\u002Fskills\u002Fdrive-nomifun","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client","https:\u002F\u002Fgithub.com\u002FGrong","2026-07-28 18:22:19",{"source":8,"skillName":5,"commands":30,"prompt":40,"zipAvailable":41},[31,34,37],{"tool":32,"command":33},"npx","npx skills add Grong\u002Fopenhub-client --skill drive-openhub",{"tool":35,"command":36},"bunx","bunx skills add Grong\u002Fopenhub-client --skill drive-openhub",{"tool":38,"command":39},"pnpm","pnpm dlx skills add Grong\u002Fopenhub-client --skill drive-openhub","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fdrive-nomifun\u002FSKILL.md",true,[],[44,51,58,65,72,79],{"skillNo":45,"name":46,"nameZh":6,"description":47,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":48,"categories":49,"keywords":50,"publishedAt":18},"01M78DF51T081P4306NA1XYJPE","example-skill","Fixture skill used by startup_materialize integration tests.","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fcrates\u002Fbackend\u002Fopenhub-extension\u002Ftests\u002Ffixtures\u002Fbuiltin-skills-fixture\u002Fexample-skill\u002FSKILL.md",[],[],{"skillNo":52,"name":53,"nameZh":6,"description":54,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":18},"01M78DF0Z8VS064W7300YJZ0KB","xiaohongshu-recruiter","用于准备在小红书上发布高质量的 AI 相关岗位招聘帖子。包含自动生成极客风格的招聘封面图和详情图，并提供确认门控的发布脚本。当用户需要发布招聘信息、寻找 Agent 设计师或其他 AI 领域人才时使用。","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fcrates\u002Fbackend\u002Fopenhub-app\u002Fassets\u002Fbuiltin-skills\u002Fxiaohongshu-recruiter\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M78DEZB3EKW0JZNZ21B2E4ZM","x-recruiter","用于在 X (x.com) 发布招聘帖子。包含文案规范、图片生成提示和自动化发布脚本。发布 AI 相关岗位或设计类岗位时优先使用。","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fcrates\u002Fbackend\u002Fopenhub-app\u002Fassets\u002Fbuiltin-skills\u002Fx-recruiter\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M78DEW7RT4CVH83HAKB5TQF5","weixin-file-send","Use when the user wants a local file or image sent back, such as \"send me the file\"\nor \"发给我\".","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fcrates\u002Fbackend\u002Fopenhub-app\u002Fassets\u002Fbuiltin-skills\u002Fweixin-file-send\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M78DERHW8WYJRWH6QCZ9HMSQ","ui-ux-pro-max","Professional UI\u002FUX design research skill with a bundled searchable database for product patterns, visual styles, color palettes, typography, landing-page structure, dashboard charts, UX practices, and stack guidance. Use when designing, building, reviewing, or improving interfaces.","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fcrates\u002Fbackend\u002Fopenhub-app\u002Fassets\u002Fbuiltin-skills\u002Fui-ux-pro-max\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M78DEKDJ3P5ZP5F8ZG894Z2T","story-roleplay","Parse and apply character cards and world info files in multiple formats (PNG, WebP, JSON), fully compatible with SillyTavern format. Supports automatic parsing, keyword triggering, and dynamic updates.","https:\u002F\u002Fgithub.com\u002FGrong\u002Fopenhub-client\u002Fblob\u002Fmain\u002Fcrates\u002Fbackend\u002Fopenhub-app\u002Fassets\u002Fbuiltin-skills\u002Fstory-roleplay\u002FSKILL.md",[],[]]