[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CAB7PSF0QXDKVZRSFSMJ9V":3,"skill-same:01M6CAB7PSF0QXDKVZRSFSMJ9V":74},{"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":42,"linkOnly":43,"markdownContent":44,"markdownBody":45,"frontMatter":46,"refName":48,"commitSha":49,"skillDirectory":50,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":6,"ownerHtmlUrl":53,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":54,"install":55,"resources":67},"01M6CAB7PSF0QXDKVZRSFSMJ9V","cmux-codex-autoapprove",null,"Run or maintain the cmux approval worker that scans cmux terminal surfaces for approval dialogs, classifies them with `codex exec`, and sends the matching approval key. Use when testing or operating the launchd-based auto-approver, debugging missed prompts, moving the worker, or tuning candidate detection and approval heuristics.","jleechanorg\u002Fjleechan-skills","jleechanorg","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F221404966?v=4","01M6C1VJ99XG950X1DWPEBCBGX","01M6C1VJDK5YVDNT3362RT5WSB",6,"MIT","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fhermes-imports\u002Fcmux-codex-autoapprove\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41],"codex","cmux codex autoapprove","cmux","when","testing","codex exec","candidate","for","operating","worker","missed","maintain","surfaces","prompts","launchd","auto","use","approval","tuning","heuristics","detection","run","debugging","terminal","2026-11-10 10:12:30",false,"---\nname: cmux-codex-autoapprove\ndescription: Run or maintain the cmux approval worker that scans cmux terminal surfaces for approval dialogs, classifies them with `codex exec`, and sends the matching approval key. Use when testing or operating the launchd-based auto-approver, debugging missed prompts, moving the worker, or tuning candidate detection and approval heuristics.\nuser-invocable: true\n---\n\n## ⚠️ Submit Discipline (MANDATORY — read this before every cmux steer)\n\n`cmux send` does **NOT** press Enter. This is the #1 recurring cmux failure mode\n(verified 2026-07-16: user explicitly flagged \"you always forget to send\" after the\nfable iOS pivot bootstrap). The **4-step ritual** below is a hard contract for every\nsend to a cmux surface. Skip ANY step and the message sits in the input buffer\nwithout ever reaching the agent.\n\n### The 4-step ritual\n\n```bash\n# STEP 1 — Type the text. OK response only proves socket acceptance, NOT submission.\ncmux send --workspace workspace:N --surface surface:M \"your message\"\n\n# STEP 2 — Press Enter. send does NOT auto-press Enter.\ncmux send-key --workspace workspace:N --surface surface:M enter\n\n# STEP 3 — Wait 5-15 seconds for the agent to start processing.\nsleep 8\n\n# STEP 4 — Verify with churning label (THE ONLY definitive proof).\ncmux capture-pane --workspace workspace:N --surface surface:M --lines 25\n# Look for one of:\n#   - \"Working (Xs • esc to interrupt)\"\n#   - \"Forming… (Xs · thinking)\"\n#   - \"Precipitating… (Xs · ↓ tokens)\"\n#   - \"Brewed \u002F Churned \u002F Cooked for Xm\"\n# If you see ANY active churning label → SUBMITTED.\n# If the text is still sitting at the ❯ prompt → NOT submitted, repeat step 2.\n# If \"Stopped\" \u002F \"Done\" \u002F nothing → no churn, investigate.\n```\n\n### ⚠️ Output Contract — typed text + terminal response (MANDATORY)\n\nEvery reply that reports a `cmux send` action MUST include, in the same reply:\n\n1. **The exact text that was typed** — verbatim copy of the string passed to `cmux send`.\n2. **The cmux terminal response** — verbatim transcript of what `cmux capture-pane` \u002F\n   `cmux read-screen` returned AFTER the `cmux send-key enter` settle window\n   (typically 5-15s). Specifically, the agent's first action after absorption.\n3. **Submission status** — explicit verdict: \"submitted (churning label X)\",\n   \"not submitted (text still at ❯ prompt)\", or \"blocked (no churn, retried N times)\".\n\n**Treat as not working until we see a response.** A reply that does NOT include\nboth the typed text AND a terminal response is invalid evidence that the\nsteer landed. The operator cannot distinguish a successful send from a failed\nsend that left text in the input buffer.\n\nCanonical contract + echo-back template: `~\u002F.hermes\u002Fskills\u002Fcmux\u002Freferences\u002Foutput-contract-mandatory.md`.\n\n### ⚠️ LLM-Provenance Caveat (MANDATORY footer)\n\nEvery reply that quotes cmux output, terminal text, or agent actions produced\nby another LLM (the worker agent OR the assistant's own synthesis of agent\noutput) MUST end with this verbatim footer:\n\n> *This was generated from another LLM and not the actual user, so feel free\n> to push back if you disagree and we can discuss.*\n\nFull caveat rules + scope: `~\u002F.hermes\u002Fskills\u002Fcmux\u002Freferences\u002Foutput-contract-mandatory.md` § \"LLM-Provenance Caveat\".\n\n### Echo-back proof (MANDATORY)\n\nEvery cmux steering action MUST be followed by an **echo-back proof** in the same\nturn or the immediate next turn to your operator (Slack thread, terminal reply,\nor whichever channel triggered the steer):\n\n> ◀ sent to surface:55 (LEFT\u002Fclaudec) at \u003CHH:MM:SS PT> — 4-step ritual complete;\n> churning label \"Forming… 9s · ↓ 4.9k tokens\" confirmed via capture-pane.\n\n**Banned** (these are the failure modes the user keeps flagging):\n- \"I sent the message\" (no Enter proof)\n- \"The agent should have received it\" (no churning label)\n- `cmux send` with no follow-up `cmux send-key enter`\n- Sending to a surface that hasn't been focused (the global focus may be on a\n  different workspace; use the raw RPC `surface.focus` if needed)\n\n### Worktree-pointer strategy for long briefs\n\nFor task briefs >200 chars (e.g. orchestrating iOS app pivot, multi-PR review),\ndo NOT paste the full text into the input. Write the brief to a file in the\nagent's cwd (e.g. `.cmux-\u003Ctask>-brief.md`) and send a 1-2 line pointer. This\navoids the autocompleter contamination pitfall where shell-style tokens inside\nlong text trigger tab completion mid-stream.\n\n### Canonical reference\n\nFull recipe + edge cases + the 2026-06-25 worked example live at:\n`~\u002F.hermes\u002Fskills\u002Fcmux\u002Freferences\u002Fsend-submit-proof-2026-06-25.md`\n\nThis rule was added 2026-07-16 after the fable iOS pivot bootstrap surfaced\n\"you always forget to send\" \u002F \"make sure you press submit and the work starts\non the cmux input\" (Slack ts 1784185650.528089). Apply it uniformly to every\ncmux-touching skill.\n\n# cmux Codex Autoapprove\n\nCanonical files:\n- Skill root: `$HOME\u002F.claude\u002Fskills\u002Fcmux-codex-autoapprove`\n- Worker script: `$HOME\u002F.claude\u002Fskills\u002Fcmux-codex-autoapprove\u002Fscripts\u002Fcmux_codex_approve_launchd.py`\n- LaunchAgent: `$HOME\u002FLibrary\u002FLaunchAgents\u002Fcom.$USER.cmux-codex-approve.plist`\n- Logs: `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd.log`\n- State: `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd-state.json`\n\nCompatibility paths:\n- Wrapper path used by older flows: `$HOME\u002F.claude\u002Fbin\u002Fcmux_codex_approve_launchd.py`\n- Codex skill symlink: `$HOME\u002F.codex\u002Fskills\u002Fcmux-codex-autoapprove`\n\n## Purpose\n\nUse this skill for a custom cmux auto-approver that:\n- enumerates terminal surfaces with `cmux --json tree --all`\n- reads visible terminal content with `cmux read-screen`\n- filters for approval-like prompts near the bottom of the live screen\n- asks `codex exec` for a one-token decision: `ENTER`, `1`, `y`, `SKIP`, or `DENY`\n- sends the chosen key back with `cmux send` or `cmux send-key`\n\nThis is the non-`snap-agent-supervisor` path.\n\n## Normal Workflow\n\n1. Verify the target surface with:\n\n```bash\ncmux --json tree --all\ncmux read-screen --workspace \u003Cworkspace> --surface \u003Csurface> --lines 24\n```\n\n2. If you need a one-shot run, execute:\n\n```bash\n\u002Fopt\u002Fhomebrew\u002Fbin\u002Fpython3 $HOME\u002F.claude\u002Fskills\u002Fcmux-codex-autoapprove\u002Fscripts\u002Fcmux_codex_approve_launchd.py\n```\n\n3. For the scheduled agent, use:\n\n```bash\nlaunchctl kickstart -k gui\u002F501\u002Fcom.$USER.cmux-codex-approve\nlaunchctl print gui\u002F501\u002Fcom.$USER.cmux-codex-approve\n```\n\n## Behavior Notes\n\n- The worker intentionally focuses on the bottom active region of the screen so stale scrollback does not retrigger approvals forever.\n- Approved prompt digests are cleared once the surface returns to a normal shell or idle prompt.\n- Idle runs are cheap. `codex exec` is only called when the screen looks like a real approval dialog.\n\n## When To Edit\n\nEdit the worker script when:\n- a real approval dialog is missed\n- a non-approval surface is being auto-approved\n- `launchd` can detect candidates but hangs or times out during classification\n- prompt wording changes and the regexes need to expand\n\nCheck these first when debugging:\n- `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd.log`\n- `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd.stderr.log`\n- `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd-state.json`\n\n","## ⚠️ Submit Discipline (MANDATORY — read this before every cmux steer)\n\n`cmux send` does **NOT** press Enter. This is the #1 recurring cmux failure mode\n(verified 2026-07-16: user explicitly flagged \"you always forget to send\" after the\nfable iOS pivot bootstrap). The **4-step ritual** below is a hard contract for every\nsend to a cmux surface. Skip ANY step and the message sits in the input buffer\nwithout ever reaching the agent.\n\n### The 4-step ritual\n\n```bash\n# STEP 1 — Type the text. OK response only proves socket acceptance, NOT submission.\ncmux send --workspace workspace:N --surface surface:M \"your message\"\n\n# STEP 2 — Press Enter. send does NOT auto-press Enter.\ncmux send-key --workspace workspace:N --surface surface:M enter\n\n# STEP 3 — Wait 5-15 seconds for the agent to start processing.\nsleep 8\n\n# STEP 4 — Verify with churning label (THE ONLY definitive proof).\ncmux capture-pane --workspace workspace:N --surface surface:M --lines 25\n# Look for one of:\n#   - \"Working (Xs • esc to interrupt)\"\n#   - \"Forming… (Xs · thinking)\"\n#   - \"Precipitating… (Xs · ↓ tokens)\"\n#   - \"Brewed \u002F Churned \u002F Cooked for Xm\"\n# If you see ANY active churning label → SUBMITTED.\n# If the text is still sitting at the ❯ prompt → NOT submitted, repeat step 2.\n# If \"Stopped\" \u002F \"Done\" \u002F nothing → no churn, investigate.\n```\n\n### ⚠️ Output Contract — typed text + terminal response (MANDATORY)\n\nEvery reply that reports a `cmux send` action MUST include, in the same reply:\n\n1. **The exact text that was typed** — verbatim copy of the string passed to `cmux send`.\n2. **The cmux terminal response** — verbatim transcript of what `cmux capture-pane` \u002F\n   `cmux read-screen` returned AFTER the `cmux send-key enter` settle window\n   (typically 5-15s). Specifically, the agent's first action after absorption.\n3. **Submission status** — explicit verdict: \"submitted (churning label X)\",\n   \"not submitted (text still at ❯ prompt)\", or \"blocked (no churn, retried N times)\".\n\n**Treat as not working until we see a response.** A reply that does NOT include\nboth the typed text AND a terminal response is invalid evidence that the\nsteer landed. The operator cannot distinguish a successful send from a failed\nsend that left text in the input buffer.\n\nCanonical contract + echo-back template: `~\u002F.hermes\u002Fskills\u002Fcmux\u002Freferences\u002Foutput-contract-mandatory.md`.\n\n### ⚠️ LLM-Provenance Caveat (MANDATORY footer)\n\nEvery reply that quotes cmux output, terminal text, or agent actions produced\nby another LLM (the worker agent OR the assistant's own synthesis of agent\noutput) MUST end with this verbatim footer:\n\n> *This was generated from another LLM and not the actual user, so feel free\n> to push back if you disagree and we can discuss.*\n\nFull caveat rules + scope: `~\u002F.hermes\u002Fskills\u002Fcmux\u002Freferences\u002Foutput-contract-mandatory.md` § \"LLM-Provenance Caveat\".\n\n### Echo-back proof (MANDATORY)\n\nEvery cmux steering action MUST be followed by an **echo-back proof** in the same\nturn or the immediate next turn to your operator (Slack thread, terminal reply,\nor whichever channel triggered the steer):\n\n> ◀ sent to surface:55 (LEFT\u002Fclaudec) at \u003CHH:MM:SS PT> — 4-step ritual complete;\n> churning label \"Forming… 9s · ↓ 4.9k tokens\" confirmed via capture-pane.\n\n**Banned** (these are the failure modes the user keeps flagging):\n- \"I sent the message\" (no Enter proof)\n- \"The agent should have received it\" (no churning label)\n- `cmux send` with no follow-up `cmux send-key enter`\n- Sending to a surface that hasn't been focused (the global focus may be on a\n  different workspace; use the raw RPC `surface.focus` if needed)\n\n### Worktree-pointer strategy for long briefs\n\nFor task briefs >200 chars (e.g. orchestrating iOS app pivot, multi-PR review),\ndo NOT paste the full text into the input. Write the brief to a file in the\nagent's cwd (e.g. `.cmux-\u003Ctask>-brief.md`) and send a 1-2 line pointer. This\navoids the autocompleter contamination pitfall where shell-style tokens inside\nlong text trigger tab completion mid-stream.\n\n### Canonical reference\n\nFull recipe + edge cases + the 2026-06-25 worked example live at:\n`~\u002F.hermes\u002Fskills\u002Fcmux\u002Freferences\u002Fsend-submit-proof-2026-06-25.md`\n\nThis rule was added 2026-07-16 after the fable iOS pivot bootstrap surfaced\n\"you always forget to send\" \u002F \"make sure you press submit and the work starts\non the cmux input\" (Slack ts 1784185650.528089). Apply it uniformly to every\ncmux-touching skill.\n\n# cmux Codex Autoapprove\n\nCanonical files:\n- Skill root: `$HOME\u002F.claude\u002Fskills\u002Fcmux-codex-autoapprove`\n- Worker script: `$HOME\u002F.claude\u002Fskills\u002Fcmux-codex-autoapprove\u002Fscripts\u002Fcmux_codex_approve_launchd.py`\n- LaunchAgent: `$HOME\u002FLibrary\u002FLaunchAgents\u002Fcom.$USER.cmux-codex-approve.plist`\n- Logs: `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd.log`\n- State: `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd-state.json`\n\nCompatibility paths:\n- Wrapper path used by older flows: `$HOME\u002F.claude\u002Fbin\u002Fcmux_codex_approve_launchd.py`\n- Codex skill symlink: `$HOME\u002F.codex\u002Fskills\u002Fcmux-codex-autoapprove`\n\n## Purpose\n\nUse this skill for a custom cmux auto-approver that:\n- enumerates terminal surfaces with `cmux --json tree --all`\n- reads visible terminal content with `cmux read-screen`\n- filters for approval-like prompts near the bottom of the live screen\n- asks `codex exec` for a one-token decision: `ENTER`, `1`, `y`, `SKIP`, or `DENY`\n- sends the chosen key back with `cmux send` or `cmux send-key`\n\nThis is the non-`snap-agent-supervisor` path.\n\n## Normal Workflow\n\n1. Verify the target surface with:\n\n```bash\ncmux --json tree --all\ncmux read-screen --workspace \u003Cworkspace> --surface \u003Csurface> --lines 24\n```\n\n2. If you need a one-shot run, execute:\n\n```bash\n\u002Fopt\u002Fhomebrew\u002Fbin\u002Fpython3 $HOME\u002F.claude\u002Fskills\u002Fcmux-codex-autoapprove\u002Fscripts\u002Fcmux_codex_approve_launchd.py\n```\n\n3. For the scheduled agent, use:\n\n```bash\nlaunchctl kickstart -k gui\u002F501\u002Fcom.$USER.cmux-codex-approve\nlaunchctl print gui\u002F501\u002Fcom.$USER.cmux-codex-approve\n```\n\n## Behavior Notes\n\n- The worker intentionally focuses on the bottom active region of the screen so stale scrollback does not retrigger approvals forever.\n- Approved prompt digests are cleared once the surface returns to a normal shell or idle prompt.\n- Idle runs are cheap. `codex exec` is only called when the screen looks like a real approval dialog.\n\n## When To Edit\n\nEdit the worker script when:\n- a real approval dialog is missed\n- a non-approval surface is being auto-approved\n- `launchd` can detect candidates but hangs or times out during classification\n- prompt wording changes and the regexes need to expand\n\nCheck these first when debugging:\n- `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd.log`\n- `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd.stderr.log`\n- `$HOME\u002F.claude\u002Fsupervisor\u002Fcmux-codex-launchd-state.json`\n\n",{"name":5,"description":7,"user-invocable":47},true,"main","05df7380981a2c4af868ce0fa57962ae0eea62a9","hermes\u002Fskills\u002Fhermes-imports\u002Fcmux-codex-autoapprove",0,"https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills","https:\u002F\u002Fgithub.com\u002Fjleechanorg","2026-11-14 03:18:13",{"source":8,"skillName":5,"commands":56,"prompt":66,"zipAvailable":47},[57,60,63],{"tool":58,"command":59},"npx","npx skills add jleechanorg\u002Fjleechan-skills --skill cmux-codex-autoapprove",{"tool":61,"command":62},"bunx","bunx skills add jleechanorg\u002Fjleechan-skills --skill cmux-codex-autoapprove",{"tool":64,"command":65},"pnpm","pnpm dlx skills add jleechanorg\u002Fjleechan-skills --skill cmux-codex-autoapprove","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fhermes-imports\u002Fcmux-codex-autoapprove\u002FSKILL.md",[68],{"path":69,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Fcmux_codex_approve_launchd.py","SCRIPT","脚本",23199,"https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fhermes-imports\u002Fcmux-codex-autoapprove\u002Fscripts\u002Fcmux_codex_approve_launchd.py",[75,91,112,126,140,155],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":42},"01M716Y5MTJ6GYJZ2ATXCJK8G4","instinct","Use when the user invokes \u002Finstinct or \u002Fin, or asks to \"message Instinct\", \"ask Instinct\", \"tell Instinct\", \"check Instinct\", or read replies from the configured peer assistant. iMessage first, WhatsApp Web as fallback.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Finstinct\u002FSKILL.md",[],[77,82,21,83,84,34,85,86,87,88,89,90],"user","replies","check","identity","whatsapp","gmail","read","message","ask",{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":102,"publishedAt":42},"01M716Y4HF7FVVMJSA2CCVN5VF","hr","Compatibility alias for history-resume; produce the same fresh-state, provenance-aware resume briefing.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhr\u002FSKILL.md",[97],{"categoryNo":98,"categoryName":99,"taxonomyType":100,"taxonomyTypeDesc":101},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[93,103,104,105,25,106,107,108,109,110,111],"resume","compatibility","provenance","history","produce","alias","briefing","state","fresh",{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":119,"publishedAt":42},"01M716Y1FFNFKFJYYKF8HQA6T8","history-resume","Resume briefing for the current repository branch or worktree using fresh state and admitted history sources.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhistory-resume\u002FSKILL.md",[118],{"categoryNo":98,"categoryName":99,"taxonomyType":100,"taxonomyTypeDesc":101},[106,103,120,121,82,122,109,123,25,124,125,110,111],"history resume","branch","analysis","repository","worktree","sources",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":42},"01M716XWXJ8EPHK4RH9T14TR85","grok","Alias for the canonical grokbot browser conversation workflow; \u002Fgrok talks to the Grok bot the same way \u002Fdot talks to Dot.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fgrok\u002FSKILL.md",[],[128,133,134,135,25,136,108,137,138,139],"dot","bot","workflow","canonical","conversation","browser","grokbot",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":42},"01M70N9Q07DTE0EW5KFXMMMSJ0","codexa-bench","Compare `codexa` and normal `codex` on a fair split of comparable work by default; use `--strict` for duplicate runs of the exact same task.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcodexa-bench\u002FSKILL.md",[],[147,148,149,34,150,25,151,152,153,154,18],"codexa bench","work","split","default","duplicate","task","compare","normal",{"skillNo":156,"name":157,"nameZh":6,"description":158,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":159,"licenseFileUrl":160,"sourceHtmlUrl":161,"categories":162,"keywords":174,"publishedAt":42},"01M6C24S2X60SP8BBFGAEKPEGW","frontend-design","Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML\u002FCSS layouts, or when styling\u002Fbeautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffrontend-design\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffrontend-design\u002FSKILL.md",[163,168,171],{"categoryNo":164,"categoryName":165,"taxonomyType":166,"taxonomyTypeDesc":167},"01JBSEEDCAT000000000000009","前端开发","SCENARIO","使用场景",{"categoryNo":169,"categoryName":170,"taxonomyType":166,"taxonomyTypeDesc":167},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":172,"categoryName":173,"taxonomyType":100,"taxonomyTypeDesc":101},"01JBSEEDCAT000000000000032","设计师",[175,176,177,34,178,179,180,82,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,21,199],"frontend","design","frontend design","generic","ui design","react","html","components","code","artifacts","ui","aesthetics","Production","web components","landing","applications","quality","web","css","create","examples","with","react components","web ui","pages"]