[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DFXVMW75D9DF65FSJ3MGDD":3,"skill-same:01M6DFXVMW75D9DF65FSJ3MGDD":72},{"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":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":34,"commitSha":51,"skillDirectory":52,"forksCount":53,"repositoryHtmlUrl":54,"repositoryHomepageUrl":6,"ownerHtmlUrl":55,"ownerBlogUrl":6,"ownerDescription":56,"updatedAt":57,"install":58,"resources":71},"01M6DFXVMW75D9DF65FSJ3MGDD","hermes-provider-oauth-operations",null,"Configure Hermes model providers, fallback chains, and OAuth re-authentication from remote or messaging sessions, especially when the `hermes` executable is not on PATH and the repo checkout must be invoked with `python -m hermes_cli.main`.","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\u002F54\u002F54741d58fbd89257cc290dabe452117a220ba248a9c8345cacf73131fb52d3d8\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"hermes","hermes provider oauth operations","provider","operations","OAuth","authentication","messaging","windows","configure","model","python","repo","auth","telegram","sessions","main","xai-oauth","openai-codex","remote-auth","linux","2026-11-10 10:12:30",false,"---\r\nname: hermes-provider-oauth-operations\r\ndescription: Configure Hermes model providers, fallback chains, and OAuth re-authentication from remote or messaging sessions, especially when the `hermes` executable is not on PATH and the repo checkout must be invoked with `python -m hermes_cli.main`.\r\nversion: 1.0.0\r\nauthor: Hermes Agent\r\nlicense: MIT\r\nplatforms: [windows, linux, macos]\r\ntags: [hermes, model, provider, oauth, xai-oauth, grok, openai-codex, telegram, remote-auth]\r\n---\r\n\r\n# Hermes Provider OAuth Operations\r\n\r\nUse this skill when the user asks to change Hermes' primary model\u002Fprovider, set a fallback\u002Fsub model, or re-authenticate an OAuth-backed provider such as `xai-oauth`\u002FGrok from Telegram or another remote messaging channel.\r\n\r\nThis skill supplements the protected `hermes-agent` skill with operational patterns learned from real remote-auth sessions.\r\n\r\n## Core workflow\r\n\r\n1. Load `hermes-agent` first for canonical Hermes commands and config semantics.\r\n2. Prefer Hermes CLI config commands over direct file edits for `~\u002F.hermes\u002Fconfig.yaml`. Do **not** assume a cached or remembered model name is still valid; verify the provider owns the model before setting it. For a cheap\u002Ffree default in this user's current setup, prefer OpenCode Zen auto-rotation:\r\n   ```bash\r\n   hermes config set model.provider opencode-zen\r\n   hermes config set model.default auto-free\r\n   hermes config set fallback_providers '[{\"provider\":\"opencode-zen\",\"model\":\"auto-free\"},{\"provider\":\"nvidia\",\"model\":\"nvidia\u002Fllama-3.1-nemotron-70b-instruct\"}]'\r\n   ```\r\n   For Codex accounts, avoid invented variants such as `gpt-5.5-low`; if using Codex, confirm the exact supported model first.\r\n3. If `hermes` is not on PATH but the repo checkout is available, run the same CLI through the module entrypoint from the repo root:\r\n   ```bash\r\n   python -m hermes_cli.main config set model.provider opencode-zen\r\n   python -m hermes_cli.main config set model.default auto-free\r\n   python -m hermes_cli.main config set fallback_providers '[{\"provider\":\"opencode-zen\",\"model\":\"auto-free\"},{\"provider\":\"nvidia\",\"model\":\"nvidia\u002Fllama-3.1-nemotron-70b-instruct\"}]'\r\n   ```\r\n4. Always verify the effective config after writes before declaring success:\r\n   ```bash\r\n   hermes config | sed -n '\u002F◆ Model\u002F,+6p'\r\n   python - \u003C\u003C'PY'\r\n   from pathlib import Path\r\n   import yaml\r\n   cfg = yaml.safe_load(Path.home().joinpath('.hermes','config.yaml').read_text(encoding='utf-8-sig'))\r\n   print(cfg.get('model', {}))\r\n   print('fallback_providers:', cfg.get('fallback_providers'))\r\n   print('reasoning_effort:', cfg.get('agent', {}).get('reasoning_effort'))\r\n   PY\r\n   ```\r\n   If a Windows atomic replace\u002Fpermission error appears after one of several `config set` calls, inspect the file before retrying: a prior command may already have left the requested provider\u002Fmodel\u002Freasoning values in place.\r\n\r\n   For this user's preferred high-end routing, do **not** put GPT5.5 under OpenCode Zen. GPT5.5 is a Codex-auth model:\r\n   ```yaml\r\n   model:\r\n     provider: openai-codex\r\n     default: gpt-5.5\r\n   fallback_providers:\r\n     - provider: openai-codex\r\n       model: gpt-5.5\r\n     - provider: nvidia\r\n       model: nvidia\u002Fllama-3.1-nemotron-70b-instruct\r\n     - provider: nous\r\n       model: nous-freetier-latest\r\n     - provider: opencode-zen\r\n       model: opencode-freetier-latest\r\n     - provider: custom\r\n       model: local-llama\r\n       base_url: http:\u002F\u002F127.0.0.1:8081\u002Fv1\r\n       key_env: LOCAL_API_KEY\r\n       start_command: powershell.exe -NoProfile -ExecutionPolicy Bypass -File \"C:\\\\Users\\\\downl\\\\Desktop\\\\hermes-agent-upstream-sync\\\\scripts\\\\windows\\\\start-hermes-llama-fallback-rtx3060.ps1\" -Port 8081 -ContextSize 65536\r\n       start_timeout_seconds: 180\r\n   ```\r\n   Leave `model.context_length` unset unless the user explicitly asks for a cap; Hermes should resolve each model's maximum context window when primary\u002Ffallback models switch.\r\n5. Start OAuth re-auth in an interactive PTY when the flow requires pasting a callback URL\u002Fcode:\r\n   ```bash\r\n   python -m hermes_cli.main auth add xai-oauth --no-browser --manual-paste --timeout 900\r\n   ```\r\n   Use `background=true` with `pty=true` so the process remains at `Callback URL:` instead of receiving EOF.\r\n6. Send the authorization URL to the user. Ask them to paste either:\r\n   - the full failed loopback callback URL, e.g. `http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fcallback?code=...&state=...`, or\r\n   - the bare authorization code if the provider displays it in-page.\r\n7. Submit the pasted callback\u002Fcode to the waiting PTY process, then verify with:\r\n   ```bash\r\n   python -m hermes_cli.main auth list xai-oauth\r\n   python -m hermes_cli.main config | sed -n '1,20p'\r\n   ```\r\n8. If this is a gateway\u002FTelegram session, tell the user whether a gateway restart or `\u002Freset` is needed for the new model\u002Fprovider state to apply to future sessions.\r\n\r\n## Pitfalls\r\n\r\n- Do not patch `~\u002F.hermes\u002Fconfig.yaml` directly when Hermes refuses sensitive config writes. Use `hermes config set` or `python -m hermes_cli.main config set`.\r\n- Non-PTY background OAuth commands can exit immediately at the paste prompt with `missing authorization code` because stdin is closed. Restart the command with `pty=true` and keep the process running.\r\n- Treat `command not found: hermes` as a PATH\u002Fsetup condition, not as a Hermes failure. Use the repo module entrypoint if available.\r\n- Do not ask the user to send tokens, refresh tokens, or secrets. The callback URL\u002Fcode is an OAuth authorization artifact for the waiting local flow; handle it only by submitting it to the active auth process.\r\n- Avoid saying re-auth is complete until `auth list` or an equivalent command shows a new usable credential.\r\n- When using the `lm-twitterer` plugin, ensure `model.provider` and `model.default` are set to values allowed by the plugin (check `plugins.entries.lm-twitterer.llm.allowed_models` and `allowed_providers` in your config).\r\n- If cron starts but job content fails with `404 Model ... not found` or `400 model is not supported`, treat it as a provider\u002Fmodel ownership mismatch until proven otherwise: inspect `agent.conversation_loop` logs, check `fallback_providers`, `delegation`, and named profile configs, then remove stale cache entries that bind model aliases to the wrong base URL.\r\n- Never put `nvidia\u002F...` model ids or `auto-free` under `provider: nous`. `auto-free` belongs to `opencode-zen`; NVIDIA NIM models belong to `provider: nvidia`; Nous fallbacks should use Nous-owned model names unless the current Nous provider catalog explicitly says otherwise.\r\n- Never put `gpt-5.5` under `provider: opencode-zen` for this user's preferred route. GPT5.5 should use `provider: openai-codex` \u002F Codex auth.\r\n- Use live free-tier sentinels (`nous-freetier-latest`, `opencode-freetier-latest`) only when the Hermes checkout expands them at runtime; otherwise update `hermes_cli\u002Ffallback_chain.py` first or use a script that refreshes the catalog before writing config.\r\n- On 429 routing requests, check whether the code retries the same credential once before falling back. If the user asks for immediate rotation, update the recovery path so rate-limit failures rotate\u002Ffallback without a same-model retry.\r\n- Local llama fallback should be cold-started via a fallback `start_command`, not launched at Hermes startup, when the user says startup is already heavy.\r\n- For NVIDIA NIM fallback in this Hermes tree, prefer the provider plugin canonical names `nvidia\u002Fllama-3.1-nemotron-70b-instruct` and `nvidia\u002Fllama-3.3-70b-instruct` over stale `nvidia\u002Fnemotron-3-*` aliases found in old catalogs\u002Fcaches.\r\n- After provider config edits, restart the gateway before judging cron behavior; the scheduler may have cached the old config in the running gateway process.\r\n\r\n## References\r\n\r\n- `references\u002Fgrok-oauth-telegram-manual-paste.md` — concise transcript-derived notes for xAI\u002FGrok OAuth via Telegram manual paste.\r\n- `references\u002Fcron-provider-model-mismatch.md` — cron\u002Ffallback debugging notes for provider\u002Fmodel mismatch, stale Nous\u002FNVIDIA aliases, and cache cleanup.\r\n- `references\u002Flive-freetier-fallback-routing.md` — Codex GPT5.5 primary, live Nous\u002FOpenCode free-tier fallback expansion, immediate 429 rotation, model-max context, and cold llama-server local fallback.\r\n","# Hermes Provider OAuth Operations\r\n\r\nUse this skill when the user asks to change Hermes' primary model\u002Fprovider, set a fallback\u002Fsub model, or re-authenticate an OAuth-backed provider such as `xai-oauth`\u002FGrok from Telegram or another remote messaging channel.\r\n\r\nThis skill supplements the protected `hermes-agent` skill with operational patterns learned from real remote-auth sessions.\r\n\r\n## Core workflow\r\n\r\n1. Load `hermes-agent` first for canonical Hermes commands and config semantics.\r\n2. Prefer Hermes CLI config commands over direct file edits for `~\u002F.hermes\u002Fconfig.yaml`. Do **not** assume a cached or remembered model name is still valid; verify the provider owns the model before setting it. For a cheap\u002Ffree default in this user's current setup, prefer OpenCode Zen auto-rotation:\r\n   ```bash\r\n   hermes config set model.provider opencode-zen\r\n   hermes config set model.default auto-free\r\n   hermes config set fallback_providers '[{\"provider\":\"opencode-zen\",\"model\":\"auto-free\"},{\"provider\":\"nvidia\",\"model\":\"nvidia\u002Fllama-3.1-nemotron-70b-instruct\"}]'\r\n   ```\r\n   For Codex accounts, avoid invented variants such as `gpt-5.5-low`; if using Codex, confirm the exact supported model first.\r\n3. If `hermes` is not on PATH but the repo checkout is available, run the same CLI through the module entrypoint from the repo root:\r\n   ```bash\r\n   python -m hermes_cli.main config set model.provider opencode-zen\r\n   python -m hermes_cli.main config set model.default auto-free\r\n   python -m hermes_cli.main config set fallback_providers '[{\"provider\":\"opencode-zen\",\"model\":\"auto-free\"},{\"provider\":\"nvidia\",\"model\":\"nvidia\u002Fllama-3.1-nemotron-70b-instruct\"}]'\r\n   ```\r\n4. Always verify the effective config after writes before declaring success:\r\n   ```bash\r\n   hermes config | sed -n '\u002F◆ Model\u002F,+6p'\r\n   python - \u003C\u003C'PY'\r\n   from pathlib import Path\r\n   import yaml\r\n   cfg = yaml.safe_load(Path.home().joinpath('.hermes','config.yaml').read_text(encoding='utf-8-sig'))\r\n   print(cfg.get('model', {}))\r\n   print('fallback_providers:', cfg.get('fallback_providers'))\r\n   print('reasoning_effort:', cfg.get('agent', {}).get('reasoning_effort'))\r\n   PY\r\n   ```\r\n   If a Windows atomic replace\u002Fpermission error appears after one of several `config set` calls, inspect the file before retrying: a prior command may already have left the requested provider\u002Fmodel\u002Freasoning values in place.\r\n\r\n   For this user's preferred high-end routing, do **not** put GPT5.5 under OpenCode Zen. GPT5.5 is a Codex-auth model:\r\n   ```yaml\r\n   model:\r\n     provider: openai-codex\r\n     default: gpt-5.5\r\n   fallback_providers:\r\n     - provider: openai-codex\r\n       model: gpt-5.5\r\n     - provider: nvidia\r\n       model: nvidia\u002Fllama-3.1-nemotron-70b-instruct\r\n     - provider: nous\r\n       model: nous-freetier-latest\r\n     - provider: opencode-zen\r\n       model: opencode-freetier-latest\r\n     - provider: custom\r\n       model: local-llama\r\n       base_url: http:\u002F\u002F127.0.0.1:8081\u002Fv1\r\n       key_env: LOCAL_API_KEY\r\n       start_command: powershell.exe -NoProfile -ExecutionPolicy Bypass -File \"C:\\\\Users\\\\downl\\\\Desktop\\\\hermes-agent-upstream-sync\\\\scripts\\\\windows\\\\start-hermes-llama-fallback-rtx3060.ps1\" -Port 8081 -ContextSize 65536\r\n       start_timeout_seconds: 180\r\n   ```\r\n   Leave `model.context_length` unset unless the user explicitly asks for a cap; Hermes should resolve each model's maximum context window when primary\u002Ffallback models switch.\r\n5. Start OAuth re-auth in an interactive PTY when the flow requires pasting a callback URL\u002Fcode:\r\n   ```bash\r\n   python -m hermes_cli.main auth add xai-oauth --no-browser --manual-paste --timeout 900\r\n   ```\r\n   Use `background=true` with `pty=true` so the process remains at `Callback URL:` instead of receiving EOF.\r\n6. Send the authorization URL to the user. Ask them to paste either:\r\n   - the full failed loopback callback URL, e.g. `http:\u002F\u002F127.0.0.1:\u003Cport>\u002Fcallback?code=...&state=...`, or\r\n   - the bare authorization code if the provider displays it in-page.\r\n7. Submit the pasted callback\u002Fcode to the waiting PTY process, then verify with:\r\n   ```bash\r\n   python -m hermes_cli.main auth list xai-oauth\r\n   python -m hermes_cli.main config | sed -n '1,20p'\r\n   ```\r\n8. If this is a gateway\u002FTelegram session, tell the user whether a gateway restart or `\u002Freset` is needed for the new model\u002Fprovider state to apply to future sessions.\r\n\r\n## Pitfalls\r\n\r\n- Do not patch `~\u002F.hermes\u002Fconfig.yaml` directly when Hermes refuses sensitive config writes. Use `hermes config set` or `python -m hermes_cli.main config set`.\r\n- Non-PTY background OAuth commands can exit immediately at the paste prompt with `missing authorization code` because stdin is closed. Restart the command with `pty=true` and keep the process running.\r\n- Treat `command not found: hermes` as a PATH\u002Fsetup condition, not as a Hermes failure. Use the repo module entrypoint if available.\r\n- Do not ask the user to send tokens, refresh tokens, or secrets. The callback URL\u002Fcode is an OAuth authorization artifact for the waiting local flow; handle it only by submitting it to the active auth process.\r\n- Avoid saying re-auth is complete until `auth list` or an equivalent command shows a new usable credential.\r\n- When using the `lm-twitterer` plugin, ensure `model.provider` and `model.default` are set to values allowed by the plugin (check `plugins.entries.lm-twitterer.llm.allowed_models` and `allowed_providers` in your config).\r\n- If cron starts but job content fails with `404 Model ... not found` or `400 model is not supported`, treat it as a provider\u002Fmodel ownership mismatch until proven otherwise: inspect `agent.conversation_loop` logs, check `fallback_providers`, `delegation`, and named profile configs, then remove stale cache entries that bind model aliases to the wrong base URL.\r\n- Never put `nvidia\u002F...` model ids or `auto-free` under `provider: nous`. `auto-free` belongs to `opencode-zen`; NVIDIA NIM models belong to `provider: nvidia`; Nous fallbacks should use Nous-owned model names unless the current Nous provider catalog explicitly says otherwise.\r\n- Never put `gpt-5.5` under `provider: opencode-zen` for this user's preferred route. GPT5.5 should use `provider: openai-codex` \u002F Codex auth.\r\n- Use live free-tier sentinels (`nous-freetier-latest`, `opencode-freetier-latest`) only when the Hermes checkout expands them at runtime; otherwise update `hermes_cli\u002Ffallback_chain.py` first or use a script that refreshes the catalog before writing config.\r\n- On 429 routing requests, check whether the code retries the same credential once before falling back. If the user asks for immediate rotation, update the recovery path so rate-limit failures rotate\u002Ffallback without a same-model retry.\r\n- Local llama fallback should be cold-started via a fallback `start_command`, not launched at Hermes startup, when the user says startup is already heavy.\r\n- For NVIDIA NIM fallback in this Hermes tree, prefer the provider plugin canonical names `nvidia\u002Fllama-3.1-nemotron-70b-instruct` and `nvidia\u002Fllama-3.3-70b-instruct` over stale `nvidia\u002Fnemotron-3-*` aliases found in old catalogs\u002Fcaches.\r\n- After provider config edits, restart the gateway before judging cron behavior; the scheduler may have cached the old config in the running gateway process.\r\n\r\n## References\r\n\r\n- `references\u002Fgrok-oauth-telegram-manual-paste.md` — concise transcript-derived notes for xAI\u002FGrok OAuth via Telegram manual paste.\r\n- `references\u002Fcron-provider-model-mismatch.md` — cron\u002Ffallback debugging notes for provider\u002Fmodel mismatch, stale Nous\u002FNVIDIA aliases, and cache cleanup.\r\n- `references\u002Flive-freetier-fallback-routing.md` — Codex GPT5.5 primary, live Nous\u002FOpenCode free-tier fallback expansion, immediate 429 rotation, model-max context, and cold llama-server local fallback.\r\n",{"name":5,"tags":44,"author":47,"license":15,"version":48,"platforms":49,"description":7},[19,28,21,45,35,46,36,32,37],"oauth","grok","Hermes Agent","1.0.0",[26,38,50],"macos","0d1434f89f4bc6b07e9d272accd8c1013768820f","archive\u002Fskill-md-reviewed\u002Fsha256\u002F54\u002F54741d58fbd89257cc290dabe452117a220ba248a9c8345cacf73131fb52d3d8",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":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add zapabob\u002FSkills --skill hermes-provider-oauth-operations",{"tool":64,"command":65},"bunx","bunx skills add zapabob\u002FSkills --skill hermes-provider-oauth-operations",{"tool":67,"command":68},"pnpm","pnpm dlx skills add zapabob\u002FSkills --skill hermes-provider-oauth-operations","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzapabob\u002FSkills\u002Fblob\u002Fmain\u002Farchive\u002Fskill-md-reviewed\u002Fsha256\u002F54\u002F54741d58fbd89257cc290dabe452117a220ba248a9c8345cacf73131fb52d3d8\u002FSKILL.md",true,[],[73,107,142,168,195,212],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":92,"publishedAt":39},"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",[79,84,87],{"categoryNo":80,"categoryName":81,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":85,"categoryName":86,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":88,"categoryName":89,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[75,93,50,94,95,38,26,96,97,98,99,100,101,19,102,103,34,104,105,106],"openclaw","cli","github","official","search","https","social-media","api","Media","hermes agent","twitter","social","via","post",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":124,"publishedAt":39},"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",[113,116,119,120,121],{"categoryNo":114,"categoryName":115,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":117,"categoryName":118,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":85,"categoryName":86,"taxonomyType":82,"taxonomyTypeDesc":83},{"categoryNo":88,"categoryName":89,"taxonomyType":90,"taxonomyTypeDesc":91},{"categoryNo":122,"categoryName":123,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000036","内容创作者",[125,126,127,128,50,129,38,130,26,131,132,133,134,135,136,137,138,139,140,141],"writing","writing plans","plans","design","review","workflow","documentation","tasks","development","code","planning","code-review","subagent","write","superpowers","paths","test-driven-development",{"skillNo":143,"name":144,"nameZh":6,"description":145,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":146,"categories":147,"keywords":151,"publishedAt":39},"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",[148],{"categoryNo":149,"categoryName":150,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000016","图像与多媒体",[144,50,38,152,153,154,155,156,157,158,28,159,160,161,162,163,164,165,166,167],"openai","speech","multimodal","Transformers","ASR","Audio","text","speech recognition","use","general","multilingual","translation","english","identification","processing","Audio Processing",{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":177,"publishedAt":39},"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",[174],{"categoryNo":175,"categoryName":176,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000013","分析与可视化",[178,50,179,180,26,181,182,183,184,185,186,28,187,188,189,190,191,192,193,194],"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":196,"name":197,"nameZh":6,"description":198,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":199,"categories":200,"keywords":202,"publishedAt":39},"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",[201],{"categoryNo":85,"categoryName":86,"taxonomyType":82,"taxonomyTypeDesc":83},[197,50,203,95,38,204,205,206,207,208,209,19,102,210,211],"terminal","automation","http","monitoring","cron","json","rss","devops","with",{"skillNo":213,"name":214,"nameZh":6,"description":215,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":216,"categories":217,"keywords":220,"publishedAt":39},"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",[218,219],{"categoryNo":80,"categoryName":81,"taxonomyType":82,"taxonomyTypeDesc":83},{"categoryNo":88,"categoryName":89,"taxonomyType":90,"taxonomyTypeDesc":91},[221,222,223,38,224,152,225,226,227,155,228,100,229,230,231,232,233,234,235,236],"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"]