[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F0916TMY0TQE81APG5DAZX":3,"skill-same:01M7F0916TMY0TQE81APG5DAZX":91},{"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":29,"commitSha":30,"skillDirectory":31,"forksCount":14,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":34,"install":35,"resources":48},"01M7F0916TMY0TQE81APG5DAZX","roast",null,"Use when someone asks to roast an idea, pressure-test or stress-test an idea, validate a business idea, \"convene the panel\", get a brutal second opinion before building something, or says \"\u002Froast\". Spins up a 5-angle panel (Critic, Champion, Analyst, Investigator, Customer) that attacks the idea from every angle, then a Judge returns one GO \u002F RESHAPE \u002F KILL verdict with the cheapest test to de-risk it.","ibada0410\u002FCLAUDE-SKILLS","ibada0410","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F237597776?v=4","01M7EQWAZNCZV6Y4AGVBXWV5CD","Ibad Ahmed","01M7EQWB08BHXD7VT182EBQDCF",0,"MIT","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: roast\ndescription: Use when someone asks to roast an idea, pressure-test or stress-test an idea, validate a business idea, \"convene the panel\", get a brutal second opinion before building something, or says \"\u002Froast\". Spins up a 5-angle panel (Critic, Champion, Analyst, Investigator, Customer) that attacks the idea from every angle, then a Judge returns one GO \u002F RESHAPE \u002F KILL verdict with the cheapest test to de-risk it.\nargument-hint: \"[the idea to roast]\"\nlicense: MIT\nmetadata:\n  version: 1.0.0\n  build_pattern: \"Path-B persona skill — adversarial panel + deterministic verdict tools\"\n  distinct_from: \"andreessen (single market-first lens, not a panel); c-level boardroom (enterprise C-suite pipeline requiring company-context onboarding); grill-me (interrogates, no verdict)\"\n---\n\n# Roast — 5-Angle Idea Panel → One Verdict\n\n> **Portability:** Reasoning-led skill with 3 stdlib Python tools. No external APIs, no LLM calls in\n> scripts. Works in Claude Code CLI and Claude.ai web. The panel does the depth; the Judge does the call.\n\n## What this does\n\nClaude's default is to agree with you. `\u002Froast` is the opposite. It convenes a panel of five\nindependent reviewers — **The Critic, The Champion, The Analyst, The Investigator, and The\nCustomer** — who tear an idea apart and build it up from every angle, then a Judge synthesizes\neverything into one honest verdict. Use it before you sink time and money into building the wrong\nthing.\n\nThe panel is adversarial on purpose. No reviewer is allowed to hedge or be polite. The point is to\nsurface what you can't see because you're too close to it.\n\n## Step 1 — Frame the idea\n\nIf `$ARGUMENTS` contains the idea, start there. Then ask the user a tight set of clarifying\nquestions so the panel has real context to work with. Ask only what hasn't already been provided.\nKeep it to 3-4 questions max, in one batch:\n\n1. **The idea** in one or two sentences (what it is, what it does).\n2. **Who it's for** and **how it makes money** (the buyer + the price\u002Fmodel).\n3. **Your edge** — relevant skills, audience, or assets you already have.\n4. **Constraints** — budget, timeline, how fast you need first dollar.\n\nIf the user says \"just run it\" or gives you enough already, skip the questions and proceed. Don't\nover-interrogate. One round, then run the panel.\n\nAssemble the brief with `scripts\u002Fbrief_builder.py` — it normalizes the four load-bearing inputs into\none paragraph and tells you if anything critical is still missing before you spend five subagents:\n\n```bash\npython scripts\u002Fbrief_builder.py \\\n  --idea \"AI that drafts grant applications for small nonprofits from a 10-min intake call\" \\\n  --who \"1-3 person nonprofits with no grant writer\" \\\n  --money \"$99\u002Fmo SaaS\" --edge \"I ran a nonprofit for 8 years\" \\\n  --constraints \"bootstrapped, first dollar in 30 days\"\n```\n\nPaste the resulting brief verbatim into every panelist's prompt, so all five judge the same thing.\n\n## Step 2 — Run the 5-angle panel (5 reviewers, in parallel)\n\nSpin up **all five reviewers in parallel in a single message** (one Task call each,\n`subagent_type: general-purpose`). Paste the same brief into each, then give each its mandate below.\n\nEach panelist must return: a one-line stance, their 3-5 sharpest points, the single most important\nthing the user must hear, and a 1-10 score on their own dimension (1 = walk away, 10 = no-brainer).\n\n**1. The Critic — \"What kills this?\"**\n> You are The Critic on an idea panel. Assume this idea fails. Your job is to find the fatal flaws, the fastest way it dies, and the load-bearing assumptions that are probably wrong. Be ruthless and specific. No hedging, no \"but it could work.\" Attack the weakest points. THE BRIEF: [brief]\n\n**2. The Champion — \"What's the 10x upside?\"**\n> You are The Champion on an idea panel. Make the strongest possible case FOR this idea. Find the biggest upside, the 10x version, the adjacent opportunities and unlock points the founder isn't seeing. Fight for the potential. Be specific about where the real money and leverage could be. THE BRIEF: [brief]\n\n**3. The Analyst — \"Does the logic actually hold?\"**\n> You are The Analyst on an idea panel. Use NO outside research and NO web. Reason purely from first principles: does the core mechanism make sense, do the incentives line up, is the underlying logic sound, does the math even work in theory? Strip it to fundamentals and tell us if it holds together. THE BRIEF: [brief]\n\n**4. The Investigator — \"What does the real market say?\"**\n> You are The Investigator on an idea panel. Use web search. Bring real-world evidence: who the existing competitors are, market size or demand signals, what comparable products charge, whether this is validated by what's already out there or contradicted by it. Cite what you find. Is the real world saying yes or no? THE BRIEF: [brief]\n\n**5. The Customer — \"Would I actually pay?\"**\n> You are The Customer on an idea panel. Role-play the exact target customer described in the brief. React as them, in first person. Would you actually pay for this? What's your real objection? What would make you choose a competitor or just do nothing instead? What price feels right, and what would make you say yes today? Be the honest, slightly skeptical customer, not a cheerleader. THE BRIEF: [brief]\n\n## Step 3 — Call the verdict\n\nOnce all five return, YOU act as the Judge. Read every panelist's findings, weigh them, and\nsynthesize one decisive verdict. **Do not just average the scores.** Run the five scores through the\nsynthesizer so the call is reproducible weighting, not vibes — then name the real tension between the\nreviewers and resolve it in prose:\n\n```bash\npython scripts\u002Fverdict_synthesizer.py \\\n  --critic 4 --champion 8 --analyst 7 --investigator 5 --customer 6\n```\n\nThe tool weights demand (Customer) and survival (Critic) heaviest and the bull (Champion) lightest,\napplies hard gates (a Customer who won't pay, or a fatal flaw the Critic landed, vetoes a GO), and\nflags the widest disagreement as the tension you must resolve. Use its verdict + confidence as your\nspine; write the prose yourself.\n\nFold in the **economics lens** yourself: rough pricing, realistic time-to-first-dollar, and whether\nthe user can actually ship this fast given the edge they described. Then design the cheapest test\nfrom the riskiest assumption the panel surfaced:\n\n```bash\npython scripts\u002Fcheapest_test_designer.py --risk price --price 99\n```\n\nOutput the verdict in this exact shape:\n\n```\n## THE VERDICT: GO \u002F RESHAPE \u002F KILL\nConfidence: [low \u002F medium \u002F high]\n\n**The call in one line:** [the decision, plainly]\n\n**Why:** [2-3 sentences resolving the panel's tension]\n\n**Biggest risk:** [the single thing most likely to kill it]\n**Biggest upside:** [the strongest reason to do it]\n\n**Money read:** [rough price, time-to-first-dollar, can they ship fast]\n\n**The cheapest 48-hour test:** [the smallest, fastest thing they can do\nto validate the riskiest assumption BEFORE building anything]\n\n**If RESHAPE:** [the specific pivot that fixes the fatal flaw while keeping the upside]\n```\n\nThen list the five panel scores in one line: `Critic X\u002F10 · Champion X\u002F10 · Analyst X\u002F10 · Investigator X\u002F10 · Customer X\u002F10`.\n\n## Tooling\n\n| Script | Role |\n|---|---|\n| `scripts\u002Fbrief_builder.py` | Normalizes the 4 load-bearing inputs into one shared brief; flags missing\u002Fthin inputs before the panel convenes. |\n| `scripts\u002Fverdict_synthesizer.py` | Weights the 5 panel scores (Customer + Critic heaviest, Champion lightest), applies veto gates, flags the real tension → GO \u002F RESHAPE \u002F KILL + confidence. |\n| `scripts\u002Fcheapest_test_designer.py` | Maps the riskiest assumption (demand\u002Fprice\u002Ffeasibility\u002Fdifferentiation\u002Fchannel\u002Fretention) to a concrete 48-hour test with pass\u002Ffail signals. |\n\n## References\n\n- [`references\u002Fadversarial_panel_canon.md`](references\u002Fadversarial_panel_canon.md) — why a diverse adversarial panel beats one reviewer (red-teaming, devil's advocacy, dialectical inquiry; 7 sources)\n- [`references\u002Fverdict_synthesis_method.md`](references\u002Fverdict_synthesis_method.md) — the weighting, the veto gates, and why you must not average (6 sources)\n- [`references\u002Fcheapest_test_canon.md`](references\u002Fcheapest_test_canon.md) — demand testing before building: smoke test, pre-sale, concierge, fake-door (7 sources)\n\n## Assets\n\n- [`assets\u002Froast_brief_worksheet.md`](assets\u002Froast_brief_worksheet.md) — fillable 4-input brief worksheet\n- [`assets\u002Fexample_roast_verdict.md`](assets\u002Fexample_roast_verdict.md) — a full worked roast (brief → 5 panel scores → tension → verdict → cheapest test)\n\n## Rules\n\n- Every reviewer stays in character. None of them hedges or softens. The value is in the friction.\n- The Judge must make an actual call. \"It depends\" is not a verdict. Pick GO, RESHAPE, or KILL and own it.\n- **Do not average the scores.** A high mean can hide a fatal split or a vetoed dimension — run the synthesizer and resolve the tension it names.\n- The cheapest 48-hour test is the most important output. It's how the user finds out if they're right without building the whole thing.\n- Keep the final verdict skimmable. The panel does the depth; the Judge does the decision.\n\n## Anti-Patterns To Reject\n\n- Softening the verdict to spare feelings (\"there's definitely something here…\"). If it's a KILL, say KILL.\n- Averaging the five scores into a mushy 6\u002F10 and calling it a day.\n- Letting the Champion's enthusiasm override a Customer who won't pay or a Critic who found the fatal flaw.\n- Ending on advice with no falsifiable test (\"go validate it\"). Name the test, the cost, and the pass\u002Ffail line.\n- Running the panel on a one-line brief so all five argue past each other.\n\n## Distinct From (don't reach for the wrong tool)\n\n- **`productivity\u002Fandreessen`** — a single market-first operator. `roast` is five independent lenses → a judge. Use andreessen when you specifically want the market-dominates thesis; use roast when you want 360° coverage.\n- **`c-level-advisor` boardroom \u002F `\u002Fcs:boardroom`** — an enterprise C-suite pipeline that needs `company-context.md` onboarding and outputs a board memo. `roast` is a zero-setup, solo-founder, 90-second gut check.\n- **`engineering\u002Fgrill-me`** — interrogates a plan one question at a time to reach shared understanding. It does not issue a GO\u002FKILL verdict. Roast judges; grill-me clarifies.\n\n---\n\n**Version:** 1.0.0\n**Build pattern:** Path-B persona skill — adversarial panel preserved + deterministic verdict tooling added.\n","# Roast — 5-Angle Idea Panel → One Verdict\n\n> **Portability:** Reasoning-led skill with 3 stdlib Python tools. No external APIs, no LLM calls in\n> scripts. Works in Claude Code CLI and Claude.ai web. The panel does the depth; the Judge does the call.\n\n## What this does\n\nClaude's default is to agree with you. `\u002Froast` is the opposite. It convenes a panel of five\nindependent reviewers — **The Critic, The Champion, The Analyst, The Investigator, and The\nCustomer** — who tear an idea apart and build it up from every angle, then a Judge synthesizes\neverything into one honest verdict. Use it before you sink time and money into building the wrong\nthing.\n\nThe panel is adversarial on purpose. No reviewer is allowed to hedge or be polite. The point is to\nsurface what you can't see because you're too close to it.\n\n## Step 1 — Frame the idea\n\nIf `$ARGUMENTS` contains the idea, start there. Then ask the user a tight set of clarifying\nquestions so the panel has real context to work with. Ask only what hasn't already been provided.\nKeep it to 3-4 questions max, in one batch:\n\n1. **The idea** in one or two sentences (what it is, what it does).\n2. **Who it's for** and **how it makes money** (the buyer + the price\u002Fmodel).\n3. **Your edge** — relevant skills, audience, or assets you already have.\n4. **Constraints** — budget, timeline, how fast you need first dollar.\n\nIf the user says \"just run it\" or gives you enough already, skip the questions and proceed. Don't\nover-interrogate. One round, then run the panel.\n\nAssemble the brief with `scripts\u002Fbrief_builder.py` — it normalizes the four load-bearing inputs into\none paragraph and tells you if anything critical is still missing before you spend five subagents:\n\n```bash\npython scripts\u002Fbrief_builder.py \\\n  --idea \"AI that drafts grant applications for small nonprofits from a 10-min intake call\" \\\n  --who \"1-3 person nonprofits with no grant writer\" \\\n  --money \"$99\u002Fmo SaaS\" --edge \"I ran a nonprofit for 8 years\" \\\n  --constraints \"bootstrapped, first dollar in 30 days\"\n```\n\nPaste the resulting brief verbatim into every panelist's prompt, so all five judge the same thing.\n\n## Step 2 — Run the 5-angle panel (5 reviewers, in parallel)\n\nSpin up **all five reviewers in parallel in a single message** (one Task call each,\n`subagent_type: general-purpose`). Paste the same brief into each, then give each its mandate below.\n\nEach panelist must return: a one-line stance, their 3-5 sharpest points, the single most important\nthing the user must hear, and a 1-10 score on their own dimension (1 = walk away, 10 = no-brainer).\n\n**1. The Critic — \"What kills this?\"**\n> You are The Critic on an idea panel. Assume this idea fails. Your job is to find the fatal flaws, the fastest way it dies, and the load-bearing assumptions that are probably wrong. Be ruthless and specific. No hedging, no \"but it could work.\" Attack the weakest points. THE BRIEF: [brief]\n\n**2. The Champion — \"What's the 10x upside?\"**\n> You are The Champion on an idea panel. Make the strongest possible case FOR this idea. Find the biggest upside, the 10x version, the adjacent opportunities and unlock points the founder isn't seeing. Fight for the potential. Be specific about where the real money and leverage could be. THE BRIEF: [brief]\n\n**3. The Analyst — \"Does the logic actually hold?\"**\n> You are The Analyst on an idea panel. Use NO outside research and NO web. Reason purely from first principles: does the core mechanism make sense, do the incentives line up, is the underlying logic sound, does the math even work in theory? Strip it to fundamentals and tell us if it holds together. THE BRIEF: [brief]\n\n**4. The Investigator — \"What does the real market say?\"**\n> You are The Investigator on an idea panel. Use web search. Bring real-world evidence: who the existing competitors are, market size or demand signals, what comparable products charge, whether this is validated by what's already out there or contradicted by it. Cite what you find. Is the real world saying yes or no? THE BRIEF: [brief]\n\n**5. The Customer — \"Would I actually pay?\"**\n> You are The Customer on an idea panel. Role-play the exact target customer described in the brief. React as them, in first person. Would you actually pay for this? What's your real objection? What would make you choose a competitor or just do nothing instead? What price feels right, and what would make you say yes today? Be the honest, slightly skeptical customer, not a cheerleader. THE BRIEF: [brief]\n\n## Step 3 — Call the verdict\n\nOnce all five return, YOU act as the Judge. Read every panelist's findings, weigh them, and\nsynthesize one decisive verdict. **Do not just average the scores.** Run the five scores through the\nsynthesizer so the call is reproducible weighting, not vibes — then name the real tension between the\nreviewers and resolve it in prose:\n\n```bash\npython scripts\u002Fverdict_synthesizer.py \\\n  --critic 4 --champion 8 --analyst 7 --investigator 5 --customer 6\n```\n\nThe tool weights demand (Customer) and survival (Critic) heaviest and the bull (Champion) lightest,\napplies hard gates (a Customer who won't pay, or a fatal flaw the Critic landed, vetoes a GO), and\nflags the widest disagreement as the tension you must resolve. Use its verdict + confidence as your\nspine; write the prose yourself.\n\nFold in the **economics lens** yourself: rough pricing, realistic time-to-first-dollar, and whether\nthe user can actually ship this fast given the edge they described. Then design the cheapest test\nfrom the riskiest assumption the panel surfaced:\n\n```bash\npython scripts\u002Fcheapest_test_designer.py --risk price --price 99\n```\n\nOutput the verdict in this exact shape:\n\n```\n## THE VERDICT: GO \u002F RESHAPE \u002F KILL\nConfidence: [low \u002F medium \u002F high]\n\n**The call in one line:** [the decision, plainly]\n\n**Why:** [2-3 sentences resolving the panel's tension]\n\n**Biggest risk:** [the single thing most likely to kill it]\n**Biggest upside:** [the strongest reason to do it]\n\n**Money read:** [rough price, time-to-first-dollar, can they ship fast]\n\n**The cheapest 48-hour test:** [the smallest, fastest thing they can do\nto validate the riskiest assumption BEFORE building anything]\n\n**If RESHAPE:** [the specific pivot that fixes the fatal flaw while keeping the upside]\n```\n\nThen list the five panel scores in one line: `Critic X\u002F10 · Champion X\u002F10 · Analyst X\u002F10 · Investigator X\u002F10 · Customer X\u002F10`.\n\n## Tooling\n\n| Script | Role |\n|---|---|\n| `scripts\u002Fbrief_builder.py` | Normalizes the 4 load-bearing inputs into one shared brief; flags missing\u002Fthin inputs before the panel convenes. |\n| `scripts\u002Fverdict_synthesizer.py` | Weights the 5 panel scores (Customer + Critic heaviest, Champion lightest), applies veto gates, flags the real tension → GO \u002F RESHAPE \u002F KILL + confidence. |\n| `scripts\u002Fcheapest_test_designer.py` | Maps the riskiest assumption (demand\u002Fprice\u002Ffeasibility\u002Fdifferentiation\u002Fchannel\u002Fretention) to a concrete 48-hour test with pass\u002Ffail signals. |\n\n## References\n\n- [`references\u002Fadversarial_panel_canon.md`](references\u002Fadversarial_panel_canon.md) — why a diverse adversarial panel beats one reviewer (red-teaming, devil's advocacy, dialectical inquiry; 7 sources)\n- [`references\u002Fverdict_synthesis_method.md`](references\u002Fverdict_synthesis_method.md) — the weighting, the veto gates, and why you must not average (6 sources)\n- [`references\u002Fcheapest_test_canon.md`](references\u002Fcheapest_test_canon.md) — demand testing before building: smoke test, pre-sale, concierge, fake-door (7 sources)\n\n## Assets\n\n- [`assets\u002Froast_brief_worksheet.md`](assets\u002Froast_brief_worksheet.md) — fillable 4-input brief worksheet\n- [`assets\u002Fexample_roast_verdict.md`](assets\u002Fexample_roast_verdict.md) — a full worked roast (brief → 5 panel scores → tension → verdict → cheapest test)\n\n## Rules\n\n- Every reviewer stays in character. None of them hedges or softens. The value is in the friction.\n- The Judge must make an actual call. \"It depends\" is not a verdict. Pick GO, RESHAPE, or KILL and own it.\n- **Do not average the scores.** A high mean can hide a fatal split or a vetoed dimension — run the synthesizer and resolve the tension it names.\n- The cheapest 48-hour test is the most important output. It's how the user finds out if they're right without building the whole thing.\n- Keep the final verdict skimmable. The panel does the depth; the Judge does the decision.\n\n## Anti-Patterns To Reject\n\n- Softening the verdict to spare feelings (\"there's definitely something here…\"). If it's a KILL, say KILL.\n- Averaging the five scores into a mushy 6\u002F10 and calling it a day.\n- Letting the Champion's enthusiasm override a Customer who won't pay or a Critic who found the fatal flaw.\n- Ending on advice with no falsifiable test (\"go validate it\"). Name the test, the cost, and the pass\u002Ffail line.\n- Running the panel on a one-line brief so all five argue past each other.\n\n## Distinct From (don't reach for the wrong tool)\n\n- **`productivity\u002Fandreessen`** — a single market-first operator. `roast` is five independent lenses → a judge. Use andreessen when you specifically want the market-dominates thesis; use roast when you want 360° coverage.\n- **`c-level-advisor` boardroom \u002F `\u002Fcs:boardroom`** — an enterprise C-suite pipeline that needs `company-context.md` onboarding and outputs a board memo. `roast` is a zero-setup, solo-founder, 90-second gut check.\n- **`engineering\u002Fgrill-me`** — interrogates a plan one question at a time to reach shared understanding. It does not issue a GO\u002FKILL verdict. Roast judges; grill-me clarifies.\n\n---\n\n**Version:** 1.0.0\n**Build pattern:** Path-B persona skill — adversarial panel preserved + deterministic verdict tooling added.\n",{"name":5,"license":15,"metadata":24,"description":7,"argument-hint":28},{"version":25,"build_pattern":26,"distinct_from":27},"1.0.0","Path-B persona skill — adversarial panel + deterministic verdict tools","andreessen (single market-first lens, not a panel); c-level boardroom (enterprise C-suite pipeline requiring company-context onboarding); grill-me (interrogates, no verdict)","[the idea to roast]","main","26a049c61e6bda33cca8f5e7c1fcdcd20681221f","Research-and-Productivity\u002Froast","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS","https:\u002F\u002Fgithub.com\u002Fibada0410","2026-10-15 17:21:48",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add ibada0410\u002FCLAUDE-SKILLS --skill roast",{"tool":41,"command":42},"bunx","bunx skills add ibada0410\u002FCLAUDE-SKILLS --skill roast",{"tool":44,"command":45},"pnpm","pnpm dlx skills add ibada0410\u002FCLAUDE-SKILLS --skill roast","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002FSKILL.md",true,[49,55,59,65,69,73,77,83,87],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"assets\u002Fexample_roast_verdict.md","ASSET","资产文件",3969,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Fassets\u002Fexample_roast_verdict.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"assets\u002Froast_brief_worksheet.md",1535,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Fassets\u002Froast_brief_worksheet.md",{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"README.md","REFERENCE","参考文档",1701,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002FREADME.md",{"path":66,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":67,"sourceHtmlUrl":68},"references\u002Fadversarial_panel_canon.md",5668,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Freferences\u002Fadversarial_panel_canon.md",{"path":70,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Fcheapest_test_canon.md",4475,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Freferences\u002Fcheapest_test_canon.md",{"path":74,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fverdict_synthesis_method.md",5336,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Freferences\u002Fverdict_synthesis_method.md",{"path":78,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":81,"sourceHtmlUrl":82},"scripts\u002Fbrief_builder.py","SCRIPT","脚本",5626,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Fscripts\u002Fbrief_builder.py",{"path":84,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":85,"sourceHtmlUrl":86},"scripts\u002Fcheapest_test_designer.py",8413,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Fscripts\u002Fcheapest_test_designer.py",{"path":88,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":89,"sourceHtmlUrl":90},"scripts\u002Fverdict_synthesizer.py",9919,"https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FResearch-and-Productivity\u002Froast\u002Fscripts\u002Fverdict_synthesizer.py",[92,101,109,117,125,133],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":96,"licenseFileUrl":97,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M7F0YW5WXY39G4W3203E8PHN","internal-comms","A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Finternal-comms\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Finternal-comms\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":105,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":19},"01M7F0YGDEVB7BCSJX8V91XK15","eu-ai-act-reviewer","Review user journeys, public content, and codebases for potentially relevant EU AI Act provisions, with exact official citations, evidence gaps, application dates, and plain-language next actions. Use for EU AI Act issue-spotting, not final legal or compliance decisions.","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Feu-ai-act-reviewer\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Feu-ai-act-reviewer\u002FSKILL.md",[],[],{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":96,"licenseFileUrl":113,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":19},"01M7F0Y1GYY34HD1WM5EWYB80Z","discernment-nudge","After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Fdiscernment-nudge\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Fdiscernment-nudge\u002FSKILL.md",[],[],{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":96,"licenseFileUrl":121,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":19},"01M7F0W4SJJY8Z345JZH9901RD","academy-guide","Stop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: \"how do I\", \"how can I\", \"getting started with\", \"what can Claude do\", \"teach me\", \"learn to use\"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Facademy-guide\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FWriting-Compliance-and-General\u002Facademy-guide\u002FSKILL.md",[],[],{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":96,"licenseFileUrl":129,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":19},"01M7F0W0X5YSTSFS77CFA5P5KJ","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FUI-and-Web-Design\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FUI-and-Web-Design\u002Fwebapp-testing\u002FSKILL.md",[],[],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":96,"licenseFileUrl":137,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":19},"01M7F0VV4P1HYHG44P0K48C4Q7","web-artifacts-builder","Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn\u002Fui). Use for complex artifacts requiring state management, routing, or shadcn\u002Fui components - not for simple single-file HTML\u002FJSX artifacts.","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FUI-and-Web-Design\u002Fweb-artifacts-builder\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fibada0410\u002FCLAUDE-SKILLS\u002Fblob\u002Fmain\u002FUI-and-Web-Design\u002Fweb-artifacts-builder\u002FSKILL.md",[],[]]