[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M795C2XSAH03PKMZX3S0FS86":3,"skill-same:01M795C2XSAH03PKMZX3S0FS86":153},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M795C2XSAH03PKMZX3S0FS86","browser-demo",null,"Run, deploy, and modify the speech-to-speech browser demo, including WebSocket\u002FWebRTC client modes, tools, search\u002Fcamera flows, Hugging Face Space limits, and UI behavior.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: browser-demo\ndescription: \"Run, deploy, and modify the speech-to-speech browser demo,\n  including WebSocket\u002FWebRTC client modes, tools, search\u002Fcamera flows, Hugging\n  Face Space limits, and UI behavior.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: Apache 2.0\n---\n\n# Browser demo\n\nUse this sub-skill when the task is to run the browser voice UI, connect it to a\n`speech-to-speech` backend, deploy it as a Space\u002Fcontainer, debug browser audio\nor transport selection, change demo tools\u002Fsearch\u002Fcamera behavior, or preserve\nUI design rules while editing the demo.\n\n## Route by task\n\n- **Start or deploy the demo server:** read\n  [Demo deployment](references\u002Fdemo-deployment.md).\n- **Debug WebSocket\u002FWebRTC browser connection, microphone, audio worklets,\n  localStorage, or Docker hostnames:** read\n  [Browser client behavior](references\u002Fbrowser-client-behavior.md).\n- **Modify web search, camera snapshot, startup greeting, account\u002Flimits, or UI\n  design:** read [Demo UI and tools](references\u002Fdemo-ui-and-tools.md).\n- **Diagnose user-facing demo failures:** read\n  [Troubleshooting](references\u002Ftroubleshooting.md).\n- **Choose backend STT\u002FLLM\u002FTTS models for the server behind the demo:** route to\n  [components-and-backends](..\u002Fcomponents-and-backends\u002FSKILL.md).\n- **Start the backend CLI or size `--num_pipelines`:** route to\n  [cli-and-server](..\u002Fcli-and-server\u002FSKILL.md).\n- **Debug raw Realtime events or WebRTC `\u002Fv1\u002Frealtime\u002Fcalls`:** route to\n  [realtime-api](..\u002Frealtime-api\u002FSKILL.md).\n\n## Minimal local sequence\n\n1. Start a backend first, normally at `ws:\u002F\u002Flocalhost:8765\u002Fv1\u002Frealtime`.\n2. Install the bundled demo server requirements in the environment that will\n   run the web app.\n3. From this sub-skill directory, set the backend URL and optional demo env vars:\n\n```bash\npython -m pip install -r assets\u002Fdemo\u002Frequirements.txt\nexport SPEECH_TO_SPEECH_URL=ws:\u002F\u002Flocalhost:8765\u002Fv1\u002Frealtime\nexport SERPER_API_KEY=...      # optional, enables server-side web search\nexport STARTUP_GREETING=...    # optional; empty disables automatic greeting\nuvicorn --app-dir assets\u002Fdemo server:app --port 7860\n```\n\n4. Open `http:\u002F\u002Flocalhost:7860\u002F`, click the orb, allow the microphone, and talk.\n   Browsers require HTTPS or localhost for microphone\u002Fcamera access.\n\n## High-signal demo rules\n\n- `SPEECH_TO_SPEECH_URL` is the highest-priority mode: the browser connects\n  directly to the pinned Realtime URL, the settings URL is locked, load-balancer\n  metering is off, and WebRTC can be offered.\n- User-entered direct URLs are WebSocket-only. WebRTC is disabled because the\n  same-origin `\u002Fapi\u002Fcalls` proxy forwards only to an env-pinned URL.\n- `LOAD_BALANCER_URL` mode is WebSocket-only and hides the backend URL from the\n  browser. When also running on a Hugging Face Space with `SPACE_ID`, OAuth\n  identity and daily time limits are active.\n- In Docker, WebSocket and WebRTC may require different backend hostnames:\n  WebSocket is dialed by the browser; WebRTC SDP is proxied by the demo server.\n- The UI color thesis is strict: saturated color belongs to the orb and tiny\n  role echoes only. Do not add colored buttons, fills, or borders outside that\n  system.\n\nThe detailed references above are self-contained so future agents can handle\nnormal operation and edits without relying on the source demo files being\npresent.\n","# Browser demo\n\nUse this sub-skill when the task is to run the browser voice UI, connect it to a\n`speech-to-speech` backend, deploy it as a Space\u002Fcontainer, debug browser audio\nor transport selection, change demo tools\u002Fsearch\u002Fcamera behavior, or preserve\nUI design rules while editing the demo.\n\n## Route by task\n\n- **Start or deploy the demo server:** read\n  [Demo deployment](references\u002Fdemo-deployment.md).\n- **Debug WebSocket\u002FWebRTC browser connection, microphone, audio worklets,\n  localStorage, or Docker hostnames:** read\n  [Browser client behavior](references\u002Fbrowser-client-behavior.md).\n- **Modify web search, camera snapshot, startup greeting, account\u002Flimits, or UI\n  design:** read [Demo UI and tools](references\u002Fdemo-ui-and-tools.md).\n- **Diagnose user-facing demo failures:** read\n  [Troubleshooting](references\u002Ftroubleshooting.md).\n- **Choose backend STT\u002FLLM\u002FTTS models for the server behind the demo:** route to\n  [components-and-backends](..\u002Fcomponents-and-backends\u002FSKILL.md).\n- **Start the backend CLI or size `--num_pipelines`:** route to\n  [cli-and-server](..\u002Fcli-and-server\u002FSKILL.md).\n- **Debug raw Realtime events or WebRTC `\u002Fv1\u002Frealtime\u002Fcalls`:** route to\n  [realtime-api](..\u002Frealtime-api\u002FSKILL.md).\n\n## Minimal local sequence\n\n1. Start a backend first, normally at `ws:\u002F\u002Flocalhost:8765\u002Fv1\u002Frealtime`.\n2. Install the bundled demo server requirements in the environment that will\n   run the web app.\n3. From this sub-skill directory, set the backend URL and optional demo env vars:\n\n```bash\npython -m pip install -r assets\u002Fdemo\u002Frequirements.txt\nexport SPEECH_TO_SPEECH_URL=ws:\u002F\u002Flocalhost:8765\u002Fv1\u002Frealtime\nexport SERPER_API_KEY=...      # optional, enables server-side web search\nexport STARTUP_GREETING=...    # optional; empty disables automatic greeting\nuvicorn --app-dir assets\u002Fdemo server:app --port 7860\n```\n\n4. Open `http:\u002F\u002Flocalhost:7860\u002F`, click the orb, allow the microphone, and talk.\n   Browsers require HTTPS or localhost for microphone\u002Fcamera access.\n\n## High-signal demo rules\n\n- `SPEECH_TO_SPEECH_URL` is the highest-priority mode: the browser connects\n  directly to the pinned Realtime URL, the settings URL is locked, load-balancer\n  metering is off, and WebRTC can be offered.\n- User-entered direct URLs are WebSocket-only. WebRTC is disabled because the\n  same-origin `\u002Fapi\u002Fcalls` proxy forwards only to an env-pinned URL.\n- `LOAD_BALANCER_URL` mode is WebSocket-only and hides the backend URL from the\n  browser. When also running on a Hugging Face Space with `SPACE_ID`, OAuth\n  identity and daily time limits are active.\n- In Docker, WebSocket and WebRTC may require different backend hostnames:\n  WebSocket is dialed by the browser; WebRTC SDP is proxied by the demo server.\n- The UI color thesis is strict: saturated color belongs to the orb and tiny\n  role echoes only. Do not add colored buttons, fills, or borders outside that\n  system.\n\nThe detailed references above are self-contained so future agents can handle\nnormal operation and edits without relying on the source demo files being\npresent.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill browser-demo",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill browser-demo",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill browser-demo","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002FSKILL.md",[47,53,59,63,67,71,75,79,83,87,91,95,99,103,107,111,115,119,123,127,131,135,141,145,149],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"assets\u002Fdemo\u002F.dockerignore","ASSET","资产文件",95,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002F.dockerignore",{"path":54,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":57,"sourceHtmlUrl":58},"assets\u002Fdemo\u002Fauth.py","SCRIPT","脚本",10539,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fauth.py",{"path":60,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":61,"sourceHtmlUrl":62},"assets\u002Fdemo\u002FDockerfile",345,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002FDockerfile",{"path":64,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":65,"sourceHtmlUrl":66},"assets\u002Fdemo\u002Fdocs\u002Fadr\u002F0001-docker-space-with-search-proxy.md",1622,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fdocs\u002Fadr\u002F0001-docker-space-with-search-proxy.md",{"path":68,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":69,"sourceHtmlUrl":70},"assets\u002Fdemo\u002Findex.html",37752,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Findex.html",{"path":72,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":73,"sourceHtmlUrl":74},"assets\u002Fdemo\u002Flimiter.py",11364,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Flimiter.py",{"path":76,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":77,"sourceHtmlUrl":78},"assets\u002Fdemo\u002Fmain.js",68631,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fmain.js",{"path":80,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":81,"sourceHtmlUrl":82},"assets\u002Fdemo\u002Frequirements.txt",166,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Frequirements.txt",{"path":84,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":85,"sourceHtmlUrl":86},"assets\u002Fdemo\u002Frtc\u002Fs2s-rtc-client.js",36991,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Frtc\u002Fs2s-rtc-client.js",{"path":88,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":89,"sourceHtmlUrl":90},"assets\u002Fdemo\u002Fserver.py",24981,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fserver.py",{"path":92,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":93,"sourceHtmlUrl":94},"assets\u002Fdemo\u002Fstyle.css",72388,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fstyle.css",{"path":96,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":97,"sourceHtmlUrl":98},"assets\u002Fdemo\u002Ftool-call-batcher.js",1935,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Ftool-call-batcher.js",{"path":100,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":101,"sourceHtmlUrl":102},"assets\u002Fdemo\u002Fui\u002Faccount.js",13523,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fui\u002Faccount.js",{"path":104,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":105,"sourceHtmlUrl":106},"assets\u002Fdemo\u002Fui\u002Fchat.js",28089,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fui\u002Fchat.js",{"path":108,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":109,"sourceHtmlUrl":110},"assets\u002Fdemo\u002Fui\u002Fdom.js",1146,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fui\u002Fdom.js",{"path":112,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":113,"sourceHtmlUrl":114},"assets\u002Fdemo\u002Fworklets\u002Faudio-playback.js",5221,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fworklets\u002Faudio-playback.js",{"path":116,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":117,"sourceHtmlUrl":118},"assets\u002Fdemo\u002Fworklets\u002Fmic-capture.js",6381,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fworklets\u002Fmic-capture.js",{"path":120,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":121,"sourceHtmlUrl":122},"assets\u002Fdemo\u002Fws\u002Fcodec.js",1866,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fws\u002Fcodec.js",{"path":124,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":125,"sourceHtmlUrl":126},"assets\u002Fdemo\u002Fws\u002Forb-visualizer.js",3658,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fws\u002Forb-visualizer.js",{"path":128,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":129,"sourceHtmlUrl":130},"assets\u002Fdemo\u002Fws\u002Fs2s-ws-client.js",49787,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fws\u002Fs2s-ws-client.js",{"path":132,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":133,"sourceHtmlUrl":134},"assets\u002Fdemo\u002Fws\u002Fuser-audio-recorder.js",7218,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Fassets\u002Fdemo\u002Fws\u002Fuser-audio-recorder.js",{"path":136,"resourceType":137,"resourceTypeDesc":138,"sizeBytes":139,"sourceHtmlUrl":140},"references\u002Fbrowser-client-behavior.md","REFERENCE","参考文档",4835,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Freferences\u002Fbrowser-client-behavior.md",{"path":142,"resourceType":137,"resourceTypeDesc":138,"sizeBytes":143,"sourceHtmlUrl":144},"references\u002Fdemo-deployment.md",5069,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Freferences\u002Fdemo-deployment.md",{"path":146,"resourceType":137,"resourceTypeDesc":138,"sizeBytes":147,"sourceHtmlUrl":148},"references\u002Fdemo-ui-and-tools.md",4917,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Freferences\u002Fdemo-ui-and-tools.md",{"path":150,"resourceType":137,"resourceTypeDesc":138,"sizeBytes":151,"sourceHtmlUrl":152},"references\u002Ftroubleshooting.md",4518,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fspeech-to-speech\u002Fsub-skills\u002Fbrowser-demo\u002Freferences\u002Ftroubleshooting.md",[154,161,168,175,182,189],{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":167,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":179,"categories":180,"keywords":181,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":188,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":190,"name":191,"nameZh":6,"description":192,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":193,"categories":194,"keywords":195,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]