[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78AX30JSZ8B6CHSRZK2NHHP":3,"skill-same:01M78AX30JSZ8B6CHSRZK2NHHP":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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":39,"commitSha":40,"skillDirectory":41,"forksCount":14,"repositoryHtmlUrl":42,"repositoryHomepageUrl":6,"ownerHtmlUrl":43,"ownerBlogUrl":6,"ownerDescription":44,"updatedAt":45,"install":46,"resources":59},"01M78AX30JSZ8B6CHSRZK2NHHP","local-voice-stack",null,"Use when installing local TTS or wiring TTS into Hermes.","abdullibrahim733-pixel\u002FEagle","abdullibrahim733-pixel","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F230504522?v=4","01M787SK0DQR0H1K326DJT6EBK","ibrahim abdull","01M787SK1HWS9V959THKZM6N70",0,"MIT","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fdevops\u002Flocal-voice-stack\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: local-voice-stack\ndescription: Use when installing local TTS or wiring TTS into Hermes.\nversion: 1.0.0\nauthor: Hermes Agent\nlicense: MIT\nplatforms: [linux]\nmetadata:\n  hermes:\n    tags: [tts, stt, voice, pocket-tts, whisper, wake-word, hermes-config, systemd]\n---\n\n# Local Voice Stack (TTS \u002F STT \u002F wake word + Hermes wiring)\n\nBuild a fully-local voice assistant on a Linux dev box: on-device neural TTS,\nspeech-to-text, wake-word detection, and — the crown jewel — wiring the TTS\nengine into Hermes as a **command TTS provider** so `text_to_speech` speaks\nthrough it. Everything runs offline, no API keys.\n\n## When to Use\n- User asks to install Pocket TTS \u002F Piper \u002F another local TTS engine.\n- User wants Hermes to *speak* through a local engine (custom TTS provider).\n- User is assembling a voice loop: wake word → STT → LLM → TTS (\"Jarvis\").\n- Any install that needs CPU-only PyTorch (avoid CUDA wheels).\n\n## Part 1 — Install a local TTS engine (Pocket TTS recipe, verified)\n\nPocket TTS = kyutai-labs\u002Fpocket-tts (`pip install pocket-tts`), 100M params,\n~6x real-time on CPU, 24 kHz mono 16-bit WAV out, voice cloning via wav file.\n\n**Recipe that works (two-step, CPU-only):**\n```bash\nuv venv --python 3.12 ~\u002F.venvs\u002Fpocket-tts\n# 1) CPU torch FIRST, from the PyTorch CPU index (no nvidia-* wheels)\n~\u002F.venvs\u002Fpocket-tts\u002Fbin\u002Fpython -m pip install --index-url https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcpu torch\n# 2) then the app, via uv (see pitfalls for why not plain pip)\nenv -u PYTHONPATH uv pip install --python ~\u002F.venvs\u002Fpocket-tts\u002Fbin\u002Fpython pocket-tts\n```\n\n**Pitfalls (each one cost a failed attempt):**\n- `uv tool install pocket-tts` resolves `torch` from PyPI → pulls GBs of\n  `nvidia-*` CUDA wheels even with no GPU. Avoid `uv tool install` here.\n- Setting `UV_INDEX_URL=https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcpu` **replaces**\n  PyPI entirely → \"pocket-tts not found in the package registry\". If you must\n  use env-var indexes, add PyPI as `UV_EXTRA_INDEX_URL`, and even then\n  first-index resolution can still grab CUDA wheels. The two-step venv recipe\n  above is the deterministic path.\n- uv-created venvs may **lack pip**: `python -m pip ...` fails with\n  \"No module named pip\" when PYTHONPATH is unset. Use\n  `uv pip install --python \u003Cvenv>\u002Fbin\u002Fpython ...` instead.\n- On this machine PYTHONPATH points at the Hermes venv → standalone venv\n  scripts import the wrong numpy (cpython-311 vs 312 crash). Always run\n  standalone venv code with `env -u PYTHONPATH` or via a wrapper script\n  (see Part 3).\n- First run downloads the model (~100M params) from Hugging Face into\n  `~\u002F.cache\u002Fhuggingface`. `HF_TOKEN` avoids rate-limit warnings.\n- CLI entrypoint lands in `\u003Cvenv>\u002Fbin\u002Fpocket-tts`; `~\u002F.local\u002Fbin` is on PATH —\n  drop a wrapper there that clears PYTHONPATH.\n\n## Part 2 — Wire into Hermes as a command TTS provider (the key technique)\n\nHermes supports user-defined TTS providers that shell out to any command.\nConfig shape (set via `hermes config set`, see pitfalls):\n\n```yaml\ntts:\n  provider: pocket\n  providers:\n    pocket:\n      type: command\n      command: \u002Fhome\u002Fcybertron\u002F.venvs\u002Fjarvis\u002Fbin\u002Fpython \u002Fhome\u002Fcybertron\u002Fjarvis\u002Fpocket_tts_cmd.py {input_path} {output_path}\n      output_format: wav\n      timeout: 60\n```\n\nContract: Hermes writes the text to `{input_path}`, runs the command; the\nadapter must write audio to `{output_path}`. Placeholders available:\n`{input_path}` (text), `{text_path}`, `{output_path}`, `{format}`, `{voice}`,\n`{model}`, `{speed}`. Valid `output_format` values: mp3, wav, ogg, flac, m4a,\naac, amr, opus. Defaults: timeout 120 s, max_text_length 5000, format mp3.\n\nSet via CLI (dotted paths work; see `references\u002Fhermes-command-tts-provider.md`\nfor source-level detail):\n```bash\nhermes config set tts.provider pocket\nhermes config set tts.providers.pocket.type command\nhermes config set tts.providers.pocket.command \"\u003Ccmd> {input_path} {output_path}\"\nhermes config set tts.providers.pocket.output_format wav\nhermes config set tts.providers.pocket.timeout 60\n```\n\n**Pitfall:** `hermes config set` prints \"not a recognized config key — saved\nanyway\" for `tts.providers.*` paths. This is **expected and harmless**:\n`tts.providers` is an open dict the tts_tool reads directly; the schema just\ndoesn't enumerate it. Don't `--force` or panic — verify by testing instead.\n\n**Verification (do this, not eyeballing config):**\n```bash\ncd ~\u002F.hermes\u002Fhermes-agent\nenv -u PYTHONPATH ~\u002F.hermes\u002Fhermes-agent\u002Fvenv\u002Fbin\u002Fpython -c \"\nfrom tools.tts_tool import text_to_speech_tool, check_tts_requirements\nprint('reqs:', check_tts_requirements())\nimport json; print(json.dumps(json.loads(text_to_speech_tool(text='test')), indent=1))\"\n# Look for: \"provider\": \"pocket\", \"success\": true, valid WAV at file_path\n```\nThen in-session: `text_to_speech` tool returns `\"provider\": \"pocket\"`.\n\n## Part 3 — Keep the server alive (systemd user service)\n\nThe TTS HTTP server dies with the terminal → make it a user service\n(no sudo needed, survives logout, auto-restarts):\n```ini\n# ~\u002F.config\u002Fsystemd\u002Fuser\u002Fpocket-tts.service\n[Unit]\nDescription=Pocket TTS server (Kyutai) - local neural TTS\nAfter=network.target\n[Service]\nType=simple\nExecStart=\u002Fhome\u002Fcybertron\u002F.local\u002Fbin\u002Fpocket-tts serve --port 8020 --quantize\nRestart=on-failure\nRestartSec=3\nEnvironment=PYTHONPATH=\n[Install]\nWantedBy=default.target\n```\n```bash\nsystemctl --user daemon-reload\nsystemctl --user enable --now pocket-tts\nsystemctl --user is-active pocket-tts && curl -sf http:\u002F\u002Flocalhost:8020\u002Fhealth\n```\n`--quantize` = int8, halves RAM. If a stray background process already holds\nthe port, kill it first or the health check may hit the wrong owner.\n\nCLI wrapper for the venv entrypoint (`~\u002F.local\u002Fbin\u002Fpocket-tts`):\n```bash\n#!\u002Fusr\u002Fbin\u002Fenv bash\nexec env -u PYTHONPATH \u002Fhome\u002Fcybertron\u002F.venvs\u002Fpocket-tts\u002Fbin\u002Fpocket-tts \"$@\"\n```\n\n## Part 4 — STT + wake word (verified legs of the loop)\n\n- **Wake word:** `openwakeword` (pip). Built-in models include `hey_jarvis`,\n  `alexa`, `timer`, `weather`. Verify:\n  `env -u PYTHONPATH \u003Cvenv>\u002Fbin\u002Fpython -c \"from openwakeword.model import Model; print(list(Model().models))\"`\n- **STT:** `faster-whisper` (pip). CPU int8: `WhisperModel('small.en', device='cpu', compute_type='int8')`.\n  Model load ~2 min on first run; transcription is fast.\n- **LLM (brain):** any OpenAI-compatible endpoint (`\u002Fv1\u002Fchat\u002Fcompletions`).\n  Note: ollama GGUFs may NOT load in upstream llama.cpp (qwen3.5 rope\n  `dimension_sections` 3 vs 4) — prefer a matching ollama runtime or a\n  compatible GGUF. (See memory for the current state of this machine's brain.)\n- Full loop scaffold lives at `~\u002Fjarvis\u002Fjarvis.py` on this machine\n  (wake → record-until-silence → whisper → LLM → Pocket TTS; `--push` mode\n  for push-to-talk). Treat it as a starting point, not gospel.\n\n## Support files\n- `templates\u002Ftts_command_adapter.py` — verified adapter: reads `{input_path}`,\n  POSTs to local TTS server, writes `{output_path}`.\n- `references\u002Fhermes-command-tts-provider.md` — source-level contract:\n  valid formats, defaults, constraints, sanitized env.\n","# Local Voice Stack (TTS \u002F STT \u002F wake word + Hermes wiring)\n\nBuild a fully-local voice assistant on a Linux dev box: on-device neural TTS,\nspeech-to-text, wake-word detection, and — the crown jewel — wiring the TTS\nengine into Hermes as a **command TTS provider** so `text_to_speech` speaks\nthrough it. Everything runs offline, no API keys.\n\n## When to Use\n- User asks to install Pocket TTS \u002F Piper \u002F another local TTS engine.\n- User wants Hermes to *speak* through a local engine (custom TTS provider).\n- User is assembling a voice loop: wake word → STT → LLM → TTS (\"Jarvis\").\n- Any install that needs CPU-only PyTorch (avoid CUDA wheels).\n\n## Part 1 — Install a local TTS engine (Pocket TTS recipe, verified)\n\nPocket TTS = kyutai-labs\u002Fpocket-tts (`pip install pocket-tts`), 100M params,\n~6x real-time on CPU, 24 kHz mono 16-bit WAV out, voice cloning via wav file.\n\n**Recipe that works (two-step, CPU-only):**\n```bash\nuv venv --python 3.12 ~\u002F.venvs\u002Fpocket-tts\n# 1) CPU torch FIRST, from the PyTorch CPU index (no nvidia-* wheels)\n~\u002F.venvs\u002Fpocket-tts\u002Fbin\u002Fpython -m pip install --index-url https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcpu torch\n# 2) then the app, via uv (see pitfalls for why not plain pip)\nenv -u PYTHONPATH uv pip install --python ~\u002F.venvs\u002Fpocket-tts\u002Fbin\u002Fpython pocket-tts\n```\n\n**Pitfalls (each one cost a failed attempt):**\n- `uv tool install pocket-tts` resolves `torch` from PyPI → pulls GBs of\n  `nvidia-*` CUDA wheels even with no GPU. Avoid `uv tool install` here.\n- Setting `UV_INDEX_URL=https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcpu` **replaces**\n  PyPI entirely → \"pocket-tts not found in the package registry\". If you must\n  use env-var indexes, add PyPI as `UV_EXTRA_INDEX_URL`, and even then\n  first-index resolution can still grab CUDA wheels. The two-step venv recipe\n  above is the deterministic path.\n- uv-created venvs may **lack pip**: `python -m pip ...` fails with\n  \"No module named pip\" when PYTHONPATH is unset. Use\n  `uv pip install --python \u003Cvenv>\u002Fbin\u002Fpython ...` instead.\n- On this machine PYTHONPATH points at the Hermes venv → standalone venv\n  scripts import the wrong numpy (cpython-311 vs 312 crash). Always run\n  standalone venv code with `env -u PYTHONPATH` or via a wrapper script\n  (see Part 3).\n- First run downloads the model (~100M params) from Hugging Face into\n  `~\u002F.cache\u002Fhuggingface`. `HF_TOKEN` avoids rate-limit warnings.\n- CLI entrypoint lands in `\u003Cvenv>\u002Fbin\u002Fpocket-tts`; `~\u002F.local\u002Fbin` is on PATH —\n  drop a wrapper there that clears PYTHONPATH.\n\n## Part 2 — Wire into Hermes as a command TTS provider (the key technique)\n\nHermes supports user-defined TTS providers that shell out to any command.\nConfig shape (set via `hermes config set`, see pitfalls):\n\n```yaml\ntts:\n  provider: pocket\n  providers:\n    pocket:\n      type: command\n      command: \u002Fhome\u002Fcybertron\u002F.venvs\u002Fjarvis\u002Fbin\u002Fpython \u002Fhome\u002Fcybertron\u002Fjarvis\u002Fpocket_tts_cmd.py {input_path} {output_path}\n      output_format: wav\n      timeout: 60\n```\n\nContract: Hermes writes the text to `{input_path}`, runs the command; the\nadapter must write audio to `{output_path}`. Placeholders available:\n`{input_path}` (text), `{text_path}`, `{output_path}`, `{format}`, `{voice}`,\n`{model}`, `{speed}`. Valid `output_format` values: mp3, wav, ogg, flac, m4a,\naac, amr, opus. Defaults: timeout 120 s, max_text_length 5000, format mp3.\n\nSet via CLI (dotted paths work; see `references\u002Fhermes-command-tts-provider.md`\nfor source-level detail):\n```bash\nhermes config set tts.provider pocket\nhermes config set tts.providers.pocket.type command\nhermes config set tts.providers.pocket.command \"\u003Ccmd> {input_path} {output_path}\"\nhermes config set tts.providers.pocket.output_format wav\nhermes config set tts.providers.pocket.timeout 60\n```\n\n**Pitfall:** `hermes config set` prints \"not a recognized config key — saved\nanyway\" for `tts.providers.*` paths. This is **expected and harmless**:\n`tts.providers` is an open dict the tts_tool reads directly; the schema just\ndoesn't enumerate it. Don't `--force` or panic — verify by testing instead.\n\n**Verification (do this, not eyeballing config):**\n```bash\ncd ~\u002F.hermes\u002Fhermes-agent\nenv -u PYTHONPATH ~\u002F.hermes\u002Fhermes-agent\u002Fvenv\u002Fbin\u002Fpython -c \"\nfrom tools.tts_tool import text_to_speech_tool, check_tts_requirements\nprint('reqs:', check_tts_requirements())\nimport json; print(json.dumps(json.loads(text_to_speech_tool(text='test')), indent=1))\"\n# Look for: \"provider\": \"pocket\", \"success\": true, valid WAV at file_path\n```\nThen in-session: `text_to_speech` tool returns `\"provider\": \"pocket\"`.\n\n## Part 3 — Keep the server alive (systemd user service)\n\nThe TTS HTTP server dies with the terminal → make it a user service\n(no sudo needed, survives logout, auto-restarts):\n```ini\n# ~\u002F.config\u002Fsystemd\u002Fuser\u002Fpocket-tts.service\n[Unit]\nDescription=Pocket TTS server (Kyutai) - local neural TTS\nAfter=network.target\n[Service]\nType=simple\nExecStart=\u002Fhome\u002Fcybertron\u002F.local\u002Fbin\u002Fpocket-tts serve --port 8020 --quantize\nRestart=on-failure\nRestartSec=3\nEnvironment=PYTHONPATH=\n[Install]\nWantedBy=default.target\n```\n```bash\nsystemctl --user daemon-reload\nsystemctl --user enable --now pocket-tts\nsystemctl --user is-active pocket-tts && curl -sf http:\u002F\u002Flocalhost:8020\u002Fhealth\n```\n`--quantize` = int8, halves RAM. If a stray background process already holds\nthe port, kill it first or the health check may hit the wrong owner.\n\nCLI wrapper for the venv entrypoint (`~\u002F.local\u002Fbin\u002Fpocket-tts`):\n```bash\n#!\u002Fusr\u002Fbin\u002Fenv bash\nexec env -u PYTHONPATH \u002Fhome\u002Fcybertron\u002F.venvs\u002Fpocket-tts\u002Fbin\u002Fpocket-tts \"$@\"\n```\n\n## Part 4 — STT + wake word (verified legs of the loop)\n\n- **Wake word:** `openwakeword` (pip). Built-in models include `hey_jarvis`,\n  `alexa`, `timer`, `weather`. Verify:\n  `env -u PYTHONPATH \u003Cvenv>\u002Fbin\u002Fpython -c \"from openwakeword.model import Model; print(list(Model().models))\"`\n- **STT:** `faster-whisper` (pip). CPU int8: `WhisperModel('small.en', device='cpu', compute_type='int8')`.\n  Model load ~2 min on first run; transcription is fast.\n- **LLM (brain):** any OpenAI-compatible endpoint (`\u002Fv1\u002Fchat\u002Fcompletions`).\n  Note: ollama GGUFs may NOT load in upstream llama.cpp (qwen3.5 rope\n  `dimension_sections` 3 vs 4) — prefer a matching ollama runtime or a\n  compatible GGUF. (See memory for the current state of this machine's brain.)\n- Full loop scaffold lives at `~\u002Fjarvis\u002Fjarvis.py` on this machine\n  (wake → record-until-silence → whisper → LLM → Pocket TTS; `--push` mode\n  for push-to-talk). Treat it as a starting point, not gospel.\n\n## Support files\n- `templates\u002Ftts_command_adapter.py` — verified adapter: reads `{input_path}`,\n  POSTs to local TTS server, writes `{output_path}`.\n- `references\u002Fhermes-command-tts-provider.md` — source-level contract:\n  valid formats, defaults, constraints, sanitized env.\n",{"name":5,"author":24,"license":15,"version":25,"metadata":26,"platforms":37,"description":7},"Hermes Agent","1.0.0",{"hermes":27},{"tags":28},[29,30,31,32,33,34,35,36],"tts","stt","voice","pocket-tts","whisper","wake-word","hermes-config","systemd",[38],"linux","main","1f420fcb3f20524e8a830b9d010e0c0ee312d437","hermes-backup\u002Fskills\u002Fdevops\u002Flocal-voice-stack","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel","I am a developer at UMG Africa and i have 2 years of software development also i engage with embedded softwares i am good in rust, cpp, python and kotlin","2026-08-16 18:04:02",{"source":8,"skillName":5,"commands":47,"prompt":57,"zipAvailable":58},[48,51,54],{"tool":49,"command":50},"npx","npx skills add abdullibrahim733-pixel\u002FEagle --skill local-voice-stack",{"tool":52,"command":53},"bunx","bunx skills add abdullibrahim733-pixel\u002FEagle --skill local-voice-stack",{"tool":55,"command":56},"pnpm","pnpm dlx skills add abdullibrahim733-pixel\u002FEagle --skill local-voice-stack","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fdevops\u002Flocal-voice-stack\u002FSKILL.md",true,[60,66],{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"references\u002Fhermes-command-tts-provider.md","REFERENCE","参考文档",3395,"https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fdevops\u002Flocal-voice-stack\u002Freferences\u002Fhermes-command-tts-provider.md",{"path":67,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":70,"sourceHtmlUrl":71},"templates\u002Ftts_command_adapter.py","SCRIPT","脚本",1057,"https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fdevops\u002Flocal-voice-stack\u002Ftemplates\u002Ftts_command_adapter.py",[73,80,87,94,101,108],{"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":79,"publishedAt":19},"01M78BBHAQCSZNG7HYMB97YG3Y","web-ui-verification","Verify web UIs via headless Chrome CDP, no vision needed.","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fsoftware-development\u002Fweb-ui-verification\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M78BB60RHXBY2QV4TA9S917B","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fsoftware-development\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M78BB16QRJZA9P3TXN5SHCQH","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fsoftware-development\u002Fsystematic-debugging\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M78BAY69B9JP3SK2NAF3EH2D","spike","Throwaway experiments to validate an idea before build.","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fsoftware-development\u002Fspike\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":19},"01M78BAVDMEBYN49PRDQ8QXTBJ","simplify-code","Parallel 4-agent cleanup of recent code changes.","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fsoftware-development\u002Fsimplify-code\u002FSKILL.md",[],[],{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":19},"01M78BAE09PQ5RV5MSJN1S788A","scroll-driven-3d-websites","Scroll-scrubbed 3D promo sites (Three.js + GSAP).","https:\u002F\u002Fgithub.com\u002Fabdullibrahim733-pixel\u002FEagle\u002Fblob\u002Fmain\u002Fhermes-backup\u002Fskills\u002Fsoftware-development\u002Fscroll-driven-3d-websites\u002FSKILL.md",[],[]]