[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6GHH0DDBFRHV20JM5KEYJ27":3,"skill-same:01M6GHH0DDBFRHV20JM5KEYJ27":83},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":28,"publishedAt":49,"linkOnly":50,"markdownContent":51,"markdownBody":52,"frontMatter":53,"refName":61,"commitSha":62,"skillDirectory":63,"forksCount":64,"repositoryHtmlUrl":65,"repositoryHomepageUrl":6,"ownerHtmlUrl":66,"ownerBlogUrl":6,"ownerDescription":67,"updatedAt":68,"install":69,"resources":82},"01M6GHH0DDBFRHV20JM5KEYJ27","desktop-dashboard-widgets",null,"Build lightweight tkinter desktop dashboards with live data, clocks, and i18n.","zapabob\u002FSkills","zapabob","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F130830376?v=4","01M6DBMYZAGRTFQDSQYW3W764P","峯岸 亮","01M6DBMZ3GKDSJ01NZ9XTZD7JD",2,"MIT","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Farchive\u002Fskill-md-reviewed\u002Fsha256\u002Ffd\u002Ffdb734985508981fa1d3cddef2e69d28db934a56da4910bc96b20cb9ba1c01f2\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000013","分析与可视化","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48],"desktop dashboard widgets","dashboard","desktop","widgets","build","with","clock","trends","computer-use","tkinter","live","i18n","macos","use","hermes agent","hermes","data","windows","linux","weather","2026-11-10 10:12:30",false,"---\r\nname: desktop-dashboard-widgets\r\ndescription: Build lightweight tkinter desktop dashboards with live data, clocks, and i18n.\r\nversion: 1.0.0\r\nauthor: Hermes Agent\r\nlicense: MIT\r\nplatforms: [windows, macos, linux]\r\nmetadata:\r\n  hermes:\r\n    tags: [tkinter, desktop, dashboard, clock, weather, trends, i18n]\r\n    category: desktop\r\n    related_skills: [computer-use]\r\n---\r\n\r\n# Desktop Dashboard Widgets Skill\r\n\r\nBuild small, always-on-top desktop dashboards with Python's standard-library tkinter. This skill covers digital and analog clocks, weather panels, public trend feeds, AI-related feeds, i18n, background refresh, and lifecycle handling. It does not replace native desktop automation; use `computer_use` for driving an existing GUI.\r\n\r\n## When to Use\r\n\r\n- The user wants a desktop clock, widget, or glanceable dashboard.\r\n- Multiple public data sources should appear in one persistent window.\r\n- The UI needs Japanese\u002FEnglish switching or another lightweight i18n layer.\r\n- The clock must follow the computer's current system time and timezone.\r\n\r\n## Prerequisites\r\n\r\n- Python with tkinter available.\r\n- On Windows, prefer the `py -3` launcher.\r\n- Public endpoints should be key-free where possible; secrets belong in `.env`, never in source.\r\n- For long-lived GUI processes, use `terminal` with `background=true`.\r\n\r\n## How to Run\r\n\r\n1. Write the application to a user-visible location such as the Desktop.\r\n2. Compile first: `py -3 -m py_compile dashboard.py`.\r\n3. Run from the script's own directory: `py -3 dashboard.py`.\r\n4. For live data, test each endpoint independently before debugging the GUI.\r\n5. Verify the process remains running and provide explicit close\u002Frestart instructions.\r\n\r\n## Quick Reference\r\n\r\n| Concern | Recommended pattern |\r\n|---|---|\r\n| System clock | `datetime.now().astimezone().tzinfo` |\r\n| UI tick | `root.after(250, update_clock)` |\r\n| Network fetch | daemon worker thread |\r\n| Worker → UI | `root.after(0, lambda: var.set(value))` |\r\n| Weather | Open-Meteo forecast endpoint |\r\n| Japan trends | Public trends page with fallback |\r\n| AI trends | Hugging Face Daily Papers API |\r\n| i18n | `TRANSLATIONS` dict + `t(key)` |\r\n| Refresh safety | Store and cancel the `after` timer id |\r\n\r\n## Procedure\r\n\r\n### 1. Establish the location and time contract\r\n\r\nAsk for the city if it is not known. For a location-specific weather panel, store city name plus latitude and longitude. For the clock, use the operating system's current time and timezone unless the user explicitly requests a fixed timezone:\r\n\r\n```python\r\nself.tz = datetime.now().astimezone().tzinfo\r\nself.local_tz_name = self.tz.tzname(datetime.now()) if self.tz else \"UTC\"\r\nnow = datetime.now(self.tz)\r\n```\r\n\r\nDo not hardcode `ZoneInfo(\"Asia\u002FTokyo\")` when the user asks to synchronize with the time currently used by the machine. Use the requested location's timezone only for the weather API query if needed.\r\n\r\n### 2. Build the window and panels\r\n\r\nUse a `Tk` root, `-topmost` for an always-on-top widget, and optional `-alpha` transparency. Use frames with stable references to each panel and label. Import tkinter geometry constants explicitly (`BOTH`, `LEFT`, `RIGHT`, `X`) rather than relying on implicit names.\r\n\r\n### 3. Implement digital and analog clocks\r\n\r\nThe digital clock should update at least once per second; 250 ms gives a smooth second transition. For a Canvas analog clock, use these angles, with `-90` degrees so 12 o'clock points upward:\r\n\r\n- Hour: `((hour % 12) + minute \u002F 60) * 30`\r\n- Minute: `(minute + second \u002F 60) * 6`\r\n- Second: `second * 6`\r\n\r\nThe hour hand is a common copy\u002Fpaste bug: it uses 30 degrees\u002Fhour, not 6.\r\n\r\n### 4. Fetch data without freezing tkinter\r\n\r\nNever perform network I\u002FO in the tkinter main loop. Start one daemon thread per independent source, catch errors, and update StringVars on the main thread:\r\n\r\n```python\r\ndef set_var(self, variable, value):\r\n    self.root.after(0, lambda: variable.set(value))\r\n\r\nthreading.Thread(target=self.fetch_weather, daemon=True).start()\r\n```\r\n\r\nSet a `User-Agent` header in urllib requests. Use bounded timeouts and show a localized fallback string if a source is unavailable.\r\n\r\n### 5. Use resilient public sources\r\n\r\nFor weather, Open-Meteo is a practical no-key source. For Japan X trends, a public trends page can be used as a best-effort fallback, but it is not equivalent to the official X API and its markup may change. For AI-related trends, Hugging Face Daily Papers provides a useful public feed of recent ML\u002FAI paper titles. Keep parsing conservative and never claim official X API data when it was not used.\r\n\r\nSee `references\u002Fdata-sources.md` for endpoints and response fields.\r\n\r\n### 6. Add i18n\r\n\r\nDefault to Japanese for this workspace. Keep visible labels and fallback messages in a `TRANSLATIONS` dictionary, expose a `t(key)` helper, and refresh panel headings when the language toggles. Avoid mixing hardcoded English into the Japanese path except for technical identifiers such as timezone names.\r\n\r\n### 7. Make refresh and shutdown safe\r\n\r\nManual refresh plus scheduled refresh can accidentally stack timers. Keep the timer id:\r\n\r\n```python\r\nif self._refresh_timer_id:\r\n    self.root.after_cancel(self._refresh_timer_id)\r\nself._refresh_timer_id = self.root.after(REFRESH_MS, self.refresh_all)\r\n```\r\n\r\nCancel the timer in `close()` before destroying the root. Bind both Escape and a visible context-menu or right-click action to close the window.\r\n\r\n### 8. Verify the real path\r\n\r\nRun syntax compilation, then run the GUI. Separately probe each endpoint and inspect parsed values. Confirm the process remains alive rather than treating a background launch with exit code 0 as proof that a GUI is visible. If GUI capture is unavailable, report that limitation honestly and provide the exact file and command for the user to inspect.\r\n\r\n## Pitfalls\r\n\r\n- **Wrong working directory on Windows shell wrappers:** run `py -3 dashboard.py` from the directory containing the file; passing an absolute path from an unrelated working directory can be mangled by a bash-style wrapper.\r\n- **Missing tkinter constants:** `RIGHT`, `BOTH`, `LEFT`, or `X` may fail at layout time with `NameError`; import them explicitly.\r\n- **Timer leaks:** do not schedule another refresh without cancelling the prior timer.\r\n- **Thread-unsafe UI updates:** worker threads must call `root.after(0, ...)`, not mutate widgets directly.\r\n- **Analog hour hand:** use 30 degrees per hour and include the minute fraction.\r\n- **Fragile trend scraping:** always provide a fallback notice and label unofficial\u002Fpublic scraping accurately.\r\n- **Dead panel variables:** store the label\u002FStringVar reference when building a panel; do not create a throwaway `StringVar` and later expect to retrieve it reliably.\r\n- **Stale location copy:** update comments, labels, and help text when changing the city; do not leave the previous city name in user-facing text.\r\n- **Language race:** workers may finish after a language toggle; keep results data-oriented and refresh after toggling, or tag requests with the active language.\r\n\r\n## Verification\r\n\r\n- `py -3 -m py_compile dashboard.py` exits successfully.\r\n- Weather endpoint returns current data for the requested coordinates.\r\n- Trend endpoint either produces a non-empty list or the localized fallback appears.\r\n- AI feed produces titles or a localized fallback appears.\r\n- Digital and analog clocks agree with the OS clock and timezone.\r\n- Manual refresh does not multiply callbacks.\r\n- Escape\u002Fright-click closes the process cleanly.\r\n- Final response includes the exact path, launch command, close method, data-source caveat, and whether live endpoint probes succeeded.\r\n\r\n## Supporting Files\r\n\r\n- `references\u002Fdata-sources.md` — endpoint contracts, parsing notes, and attribution caveats.\r\n- `templates\u002Ftkinter-dashboard.py` — compact starter skeleton for future dashboards.\r\n","# Desktop Dashboard Widgets Skill\r\n\r\nBuild small, always-on-top desktop dashboards with Python's standard-library tkinter. This skill covers digital and analog clocks, weather panels, public trend feeds, AI-related feeds, i18n, background refresh, and lifecycle handling. It does not replace native desktop automation; use `computer_use` for driving an existing GUI.\r\n\r\n## When to Use\r\n\r\n- The user wants a desktop clock, widget, or glanceable dashboard.\r\n- Multiple public data sources should appear in one persistent window.\r\n- The UI needs Japanese\u002FEnglish switching or another lightweight i18n layer.\r\n- The clock must follow the computer's current system time and timezone.\r\n\r\n## Prerequisites\r\n\r\n- Python with tkinter available.\r\n- On Windows, prefer the `py -3` launcher.\r\n- Public endpoints should be key-free where possible; secrets belong in `.env`, never in source.\r\n- For long-lived GUI processes, use `terminal` with `background=true`.\r\n\r\n## How to Run\r\n\r\n1. Write the application to a user-visible location such as the Desktop.\r\n2. Compile first: `py -3 -m py_compile dashboard.py`.\r\n3. Run from the script's own directory: `py -3 dashboard.py`.\r\n4. For live data, test each endpoint independently before debugging the GUI.\r\n5. Verify the process remains running and provide explicit close\u002Frestart instructions.\r\n\r\n## Quick Reference\r\n\r\n| Concern | Recommended pattern |\r\n|---|---|\r\n| System clock | `datetime.now().astimezone().tzinfo` |\r\n| UI tick | `root.after(250, update_clock)` |\r\n| Network fetch | daemon worker thread |\r\n| Worker → UI | `root.after(0, lambda: var.set(value))` |\r\n| Weather | Open-Meteo forecast endpoint |\r\n| Japan trends | Public trends page with fallback |\r\n| AI trends | Hugging Face Daily Papers API |\r\n| i18n | `TRANSLATIONS` dict + `t(key)` |\r\n| Refresh safety | Store and cancel the `after` timer id |\r\n\r\n## Procedure\r\n\r\n### 1. Establish the location and time contract\r\n\r\nAsk for the city if it is not known. For a location-specific weather panel, store city name plus latitude and longitude. For the clock, use the operating system's current time and timezone unless the user explicitly requests a fixed timezone:\r\n\r\n```python\r\nself.tz = datetime.now().astimezone().tzinfo\r\nself.local_tz_name = self.tz.tzname(datetime.now()) if self.tz else \"UTC\"\r\nnow = datetime.now(self.tz)\r\n```\r\n\r\nDo not hardcode `ZoneInfo(\"Asia\u002FTokyo\")` when the user asks to synchronize with the time currently used by the machine. Use the requested location's timezone only for the weather API query if needed.\r\n\r\n### 2. Build the window and panels\r\n\r\nUse a `Tk` root, `-topmost` for an always-on-top widget, and optional `-alpha` transparency. Use frames with stable references to each panel and label. Import tkinter geometry constants explicitly (`BOTH`, `LEFT`, `RIGHT`, `X`) rather than relying on implicit names.\r\n\r\n### 3. Implement digital and analog clocks\r\n\r\nThe digital clock should update at least once per second; 250 ms gives a smooth second transition. For a Canvas analog clock, use these angles, with `-90` degrees so 12 o'clock points upward:\r\n\r\n- Hour: `((hour % 12) + minute \u002F 60) * 30`\r\n- Minute: `(minute + second \u002F 60) * 6`\r\n- Second: `second * 6`\r\n\r\nThe hour hand is a common copy\u002Fpaste bug: it uses 30 degrees\u002Fhour, not 6.\r\n\r\n### 4. Fetch data without freezing tkinter\r\n\r\nNever perform network I\u002FO in the tkinter main loop. Start one daemon thread per independent source, catch errors, and update StringVars on the main thread:\r\n\r\n```python\r\ndef set_var(self, variable, value):\r\n    self.root.after(0, lambda: variable.set(value))\r\n\r\nthreading.Thread(target=self.fetch_weather, daemon=True).start()\r\n```\r\n\r\nSet a `User-Agent` header in urllib requests. Use bounded timeouts and show a localized fallback string if a source is unavailable.\r\n\r\n### 5. Use resilient public sources\r\n\r\nFor weather, Open-Meteo is a practical no-key source. For Japan X trends, a public trends page can be used as a best-effort fallback, but it is not equivalent to the official X API and its markup may change. For AI-related trends, Hugging Face Daily Papers provides a useful public feed of recent ML\u002FAI paper titles. Keep parsing conservative and never claim official X API data when it was not used.\r\n\r\nSee `references\u002Fdata-sources.md` for endpoints and response fields.\r\n\r\n### 6. Add i18n\r\n\r\nDefault to Japanese for this workspace. Keep visible labels and fallback messages in a `TRANSLATIONS` dictionary, expose a `t(key)` helper, and refresh panel headings when the language toggles. Avoid mixing hardcoded English into the Japanese path except for technical identifiers such as timezone names.\r\n\r\n### 7. Make refresh and shutdown safe\r\n\r\nManual refresh plus scheduled refresh can accidentally stack timers. Keep the timer id:\r\n\r\n```python\r\nif self._refresh_timer_id:\r\n    self.root.after_cancel(self._refresh_timer_id)\r\nself._refresh_timer_id = self.root.after(REFRESH_MS, self.refresh_all)\r\n```\r\n\r\nCancel the timer in `close()` before destroying the root. Bind both Escape and a visible context-menu or right-click action to close the window.\r\n\r\n### 8. Verify the real path\r\n\r\nRun syntax compilation, then run the GUI. Separately probe each endpoint and inspect parsed values. Confirm the process remains alive rather than treating a background launch with exit code 0 as proof that a GUI is visible. If GUI capture is unavailable, report that limitation honestly and provide the exact file and command for the user to inspect.\r\n\r\n## Pitfalls\r\n\r\n- **Wrong working directory on Windows shell wrappers:** run `py -3 dashboard.py` from the directory containing the file; passing an absolute path from an unrelated working directory can be mangled by a bash-style wrapper.\r\n- **Missing tkinter constants:** `RIGHT`, `BOTH`, `LEFT`, or `X` may fail at layout time with `NameError`; import them explicitly.\r\n- **Timer leaks:** do not schedule another refresh without cancelling the prior timer.\r\n- **Thread-unsafe UI updates:** worker threads must call `root.after(0, ...)`, not mutate widgets directly.\r\n- **Analog hour hand:** use 30 degrees per hour and include the minute fraction.\r\n- **Fragile trend scraping:** always provide a fallback notice and label unofficial\u002Fpublic scraping accurately.\r\n- **Dead panel variables:** store the label\u002FStringVar reference when building a panel; do not create a throwaway `StringVar` and later expect to retrieve it reliably.\r\n- **Stale location copy:** update comments, labels, and help text when changing the city; do not leave the previous city name in user-facing text.\r\n- **Language race:** workers may finish after a language toggle; keep results data-oriented and refresh after toggling, or tag requests with the active language.\r\n\r\n## Verification\r\n\r\n- `py -3 -m py_compile dashboard.py` exits successfully.\r\n- Weather endpoint returns current data for the requested coordinates.\r\n- Trend endpoint either produces a non-empty list or the localized fallback appears.\r\n- AI feed produces titles or a localized fallback appears.\r\n- Digital and analog clocks agree with the OS clock and timezone.\r\n- Manual refresh does not multiply callbacks.\r\n- Escape\u002Fright-click closes the process cleanly.\r\n- Final response includes the exact path, launch command, close method, data-source caveat, and whether live endpoint probes succeeded.\r\n\r\n## Supporting Files\r\n\r\n- `references\u002Fdata-sources.md` — endpoint contracts, parsing notes, and attribution caveats.\r\n- `templates\u002Ftkinter-dashboard.py` — compact starter skeleton for future dashboards.\r\n",{"name":5,"author":54,"license":15,"version":55,"metadata":56,"platforms":60,"description":7},"Hermes Agent","1.0.0",{"hermes":57},{"tags":58,"category":31,"related_skills":59},[38,31,30,35,48,36,40],[37],[46,41,47],"main","0d1434f89f4bc6b07e9d272accd8c1013768820f","archive\u002Fskill-md-reviewed\u002Fsha256\u002Ffd\u002Ffdb734985508981fa1d3cddef2e69d28db934a56da4910bc96b20cb9ba1c01f2",0,"https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills","https:\u002F\u002Fgithub.com\u002Fzapabob","The Open University of JAPAN　2019-now\r\n\r\nTokyo University of Science　2007-2010","2026-10-26 16:29:01",{"source":8,"skillName":5,"commands":70,"prompt":80,"zipAvailable":81},[71,74,77],{"tool":72,"command":73},"npx","npx skills add zapabob\u002FSkills --skill desktop-dashboard-widgets",{"tool":75,"command":76},"bunx","bunx skills add zapabob\u002FSkills --skill desktop-dashboard-widgets",{"tool":78,"command":79},"pnpm","pnpm dlx skills add zapabob\u002FSkills --skill desktop-dashboard-widgets","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Farchive\u002Fskill-md-reviewed\u002Fsha256\u002Ffd\u002Ffdb734985508981fa1d3cddef2e69d28db934a56da4910bc96b20cb9ba1c01f2\u002FSKILL.md",true,[],[84,113,148,174,199,215],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":99,"publishedAt":49},"01M6GJTE1GESGGF9K028Y7753E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Fregistry\u002Fskills\u002Fxurl\u002Fvariants\u002Fhermes-agent\u002FSKILL.md",[90,93,96],{"categoryNo":91,"categoryName":92,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":94,"categoryName":95,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":97,"categoryName":98,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000030","开发工程师",[86,100,41,101,102,47,46,103,104,105,106,107,108,44,43,109,61,110,111,112],"openclaw","cli","github","official","search","https","social-media","api","Media","twitter","social","via","post",{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":130,"publishedAt":49},"01M6GJT8S1PXQ8WPXT1HSNBWH2","writing-plans","Write implementation plans: bite-sized tasks, paths, code.","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Fregistry\u002Fskills\u002Fwriting-plans\u002Fvariants\u002Fhermes-agent\u002FSKILL.md",[119,122,125,126,127],{"categoryNo":120,"categoryName":121,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":123,"categoryName":124,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":94,"categoryName":95,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":97,"categoryName":98,"taxonomyType":26,"taxonomyTypeDesc":27},{"categoryNo":128,"categoryName":129,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000036","内容创作者",[131,132,133,134,41,135,47,136,46,137,138,139,140,141,142,143,144,145,146,147],"writing","writing plans","plans","design","review","workflow","documentation","tasks","development","code","planning","code-review","subagent","write","superpowers","paths","test-driven-development",{"skillNo":149,"name":150,"nameZh":6,"description":151,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":157,"publishedAt":49},"01M6GJST0H5X7WHBJ3BZK0DVKH","whisper","OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Fregistry\u002Fskills\u002Fwhisper\u002Fvariants\u002Fhermes-agent-optional\u002FSKILL.md",[154],{"categoryNo":155,"categoryName":156,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",[150,41,47,158,159,160,161,162,163,164,165,166,42,167,168,169,170,171,172,173],"openai","speech","multimodal","Transformers","ASR","Audio","text","model","speech recognition","general","multilingual","translation","english","identification","processing","Audio Processing",{"skillNo":175,"name":176,"nameZh":6,"description":177,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":178,"categories":179,"keywords":181,"publishedAt":49},"01M6GJSMRQ8EETZTYMG72P70EA","weights-and-biases","W&B: log ML experiments, sweeps, model registry, dashboards.","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Fregistry\u002Fskills\u002Fweights-and-biases\u002Fvariants\u002Fhermes-agent\u002FSKILL.md",[180],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[182,41,183,184,46,185,186,187,188,189,190,165,191,192,193,194,195,196,197,198],"weights and biases","tensorflow","experiments","pytorch","visualization","registry","real-time","tuning","dashboards","experiment","log","time","Experiment Tracking","Model Registry","Hyperparameter Tuning","WandB","Real-Time Visualization",{"skillNo":200,"name":201,"nameZh":6,"description":202,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":203,"categories":204,"keywords":206,"publishedAt":49},"01M6GJS19KDPRY112JTR326S06","watchers","Poll RSS, JSON APIs, and GitHub with watermark dedup.","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Fregistry\u002Fskills\u002Fwatchers\u002Fvariants\u002Fhermes-agent-optional\u002FSKILL.md",[205],{"categoryNo":94,"categoryName":95,"taxonomyType":21,"taxonomyTypeDesc":22},[201,41,207,102,47,208,209,210,211,212,213,44,43,214,34],"terminal","automation","http","monitoring","cron","json","rss","devops",{"skillNo":216,"name":217,"nameZh":6,"description":218,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":219,"categories":220,"keywords":223,"publishedAt":49},"01M6GJRHJD2A6Q830SWF9W48BC","serving-llms-vllm","vLLM: high-throughput LLM serving, OpenAI API, quantization.","https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Fregistry\u002Fskills\u002Fvllm\u002Fvariants\u002Fhermes-agent\u002FSKILL.md",[221,222],{"categoryNo":91,"categoryName":92,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":97,"categoryName":98,"taxonomyType":26,"taxonomyTypeDesc":27},[224,225,226,47,227,158,228,229,230,161,231,107,232,233,234,235,236,237,238,239],"llms","vLLM","serving llms vllm","research","openai api","tensor","Inference Serving","Production","llm serving","High Throughput","PagedAttention","Continuous Batching","Quantization","Tensor Parallelism","high-throughput","throughput"]