[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DB8Z79DSTJZB11PMWP6GXA":3,"skill-same:01M6DB8Z79DSTJZB11PMWP6GXA":116},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":28,"publishedAt":49,"linkOnly":50,"markdownContent":51,"markdownBody":52,"frontMatter":53,"refName":56,"commitSha":57,"skillDirectory":58,"forksCount":14,"repositoryHtmlUrl":59,"repositoryHomepageUrl":6,"ownerHtmlUrl":60,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":61,"install":62,"resources":75},"01M6DB8Z79DSTJZB11PMWP6GXA","critique-panel",null,"Use to run the G3 CRITIQUE gate of the TCR design-review loop - a four named-persona panel (user-advocate \u002F systems-consistency \u002F a11y \u002F brand-aesthetic) that each REASON over ui-audit's Track-A numbers plus the screenshot, run the NN\u002Fg \"I Like \u002F I Wish \u002F What If\" ritual, score the nine DESIGN-EVAL v1 dimensions 0-4, then synthesize a severity-ranked findings list. A TRIAGE accelerator.","Prekzursil\u002Fagent-skills-toolchain","Prekzursil","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54636077?v=4","01M6DACZ1PCQHD99SCR1YXG28B","Andrei Mihai Visalon","01M6DACZ670WZF535V3WYYSS4Z",0,"MIT","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48],"critique panel","critique","brand","systems","synthesize","dimensions","gate","aesthetic","use","list","eval","design","loop","persona","severity","audit","screenshot","run","triage","review","2026-11-10 10:12:30",false,"---\nname: critique-panel\ndescription: Use to run the G3 CRITIQUE gate of the TCR design-review loop - a four named-persona panel (user-advocate \u002F systems-consistency \u002F a11y \u002F brand-aesthetic) that each REASON over ui-audit's Track-A numbers plus the screenshot, run the NN\u002Fg \"I Like \u002F I Wish \u002F What If\" ritual, score the nine DESIGN-EVAL v1 dimensions 0-4, then synthesize a severity-ranked findings list. A TRIAGE accelerator.\nmetadata:\n  origin: user\n---\n\n\u003C!-- origin: user (design-SOTA research, 2026-06-30); implements DESIGN-WORKFLOW.md G3 (NN\u002Fg Design Critique ritual + I Like\u002FI Wish\u002FWhat If + named multi-agent panel as TRIAGE), DESIGN-SKILLS-AND-EVAL.md §1.1 (one-persona-per-call reliability protocol), §1.2 (the four personas + quota), §2.1-§2.4 (0-4 scale, 9 dimensions\u002Fweights\u002Fcritical triggers, weighted composite + cap), §3 (defer DIRECTION to the human; enumerate don't pick). -->\n\n\n# critique-panel Skill\n\nThe **G3 CRITIQUE** gate of the TCR design-review loop — the *discuss edge*. Four named\nheuristic personas each reason over the **same** pre-computed Track-A numbers\n(`ui-audit` `audit.json`) + the screenshot, run the NN\u002Fg design-critique ritual, score\nthe nine **DESIGN-EVAL v1** dimensions 0–4, and emit a typed verdict. `synthesize` then\ndedupes + ranks them into one findings list with a weighted composite.\n\n**Skill location:** `~\u002F.claude\u002Fskills\u002Fcritique-panel\u002F`\n\n> The harness (`critique_panel.mjs`) is **pure plumbing** — it never calls a model and\n> never measures pixels. It routes evidence onto each persona's owned dimensions, builds\n> the per-persona brief, validates typed verdicts (incl. the quota gate), and aggregates.\n\n---\n\n## When to use\n\n- The **G3 CRITIQUE** gate after `ui-audit` (G1) has produced `audit.json` + screenshots.\n- You want a **multi-perspective**, severity-rated, typed critique (not \"looks good to me\").\n- You want the panel to **enumerate + rank** defects and propose directions for the human.\n\n**When NOT to use:**\n- Driving the app \u002F measuring pixels → that is `ui-audit` Track-A (this consumes its output).\n- Picking the winning redesign direction → the **human decides at G4** (`frontend-design-direction`).\n- The whole 8-gate loop end-to-end → `design-studio` (this is one leaf gate).\n\n---\n\n## The honesty note (cite it)\n\nThe panel is a **TRIAGE accelerator, NOT an oracle.** Multi-agent VLM critique has ~21%\nexpert overlap, ~24% false-positives, and is near-random at *picking* the better design\n(WiserUI-Bench). So the personas **enumerate, score, and rank** — the **human is the\ndecider at G4** (`DESIGN-SKILLS-AND-EVAL.md §3.2`).\n\n---\n\n## The four named personas\n\nEach persona stays in character and judges **only** its owned dimensions (one persona per\ncall kills the halo effect — `DESIGN-SKILLS-AND-EVAL.md §1.1`). Dimensions are\ndouble-covered on the high-weight \u002F critical-prone axes so the panel cross-checks.\n\n| Persona | Method | Owned dimensions |\n|---|---|---|\n| **user-advocate** | Nielsen's 10 usability heuristics | visual-hierarchy, empty-error-states, affordance, consistency |\n| **systems-consistency** | Refactoring UI + Material 3 \u002F HIG + Nielsen #4 | spacing-rhythm, type-scale, alignment, consistency |\n| **a11y** | WCAG 2.2 AA + WCAG 3.0 draft | color-contrast, affordance, motion |\n| **brand-aesthetic** | Refactoring UI colour\u002Fpersonality + Nielsen #8 | visual-hierarchy, color-contrast, type-scale, motion |\n\nThe full per-persona method + the 0–4 band, severity ladder, nine dimensions\u002Fweights, and\ncritical-error triggers live in **`panel_rubric.md`** (the deterministic lens, schema-aligned).\n\n---\n\n## Driving the panel (the harness)\n\n```powershell\n$CP = \"$HOME\u002F.claude\u002Fskills\u002Fcritique-panel\u002Fcritique_panel.mjs\"\n\n# 1) ASSEMBLE — read audit.json + screenshot(s) + rubric, route numbers onto each\n#    persona's owned dimensions, write FOUR persona briefs (.json + .md) to fill.\nnode $CP assemble --audit .\u002Fout\u002Faudit.json --screenshot a.png,b.png --scope \u002F,\u002Fsettings --out .\u002Fout\n\n# 2) (Claude, as each persona) fill panel-\u003Cpersona>.out.json — score 0-4 + I Like\u002FI Wish\u002FWhat If.\n\n# 3) VALIDATE — type-check a persona verdict incl. the quota gate (>=2 iLike + 1 iWish\u002Fscreen)\nnode $CP validate --input .\u002Fout\u002Fpanel-user-advocate.out.json\n#   -> SUCCESS:critique-panel-user-advocate-\u003Cid> \u002F FAILED:...\n\n# 4) SYNTHESIZE — merge verdicts -> ranked findings + weighted composite + protected wins\nnode $CP synthesize --inputs .\u002Fout\u002Fpanel-*.out.json --audit .\u002Fout\u002Faudit.json --out .\u002Fout\n#   -> SUCCESS:critique-panel-\u003Cid>\n\n# Dump the persona + synthesis JSON Schemas (keep them in sync)\nnode $CP schema --out .\u002F\n```\n\n**The contract:** the harness routes evidence and validates; **Claude reasons** as each\npersona over the numbers; `synthesize` computes the verdict. The personas reason over\nnumbers Track-A already measured — they never pixel-ground.\n\n---\n\n## Scoring (from `panel_rubric.md`)\n\n- **0–4 per owned dimension** + a `confidence`; `critical: true` only when that dimension's\n  critical-error trigger fired (forces its score to **0**).\n- **I Like** (wins to protect) \u002F **I Wish** (severity-rated defects + recommendation) \u002F\n  **What If** (alternative directions to enumerate — *not* pick).\n- **Quota (anti-tunnel-vision, even solo):** ≥2 `iLike` + 1 `iWish` per in-scope screen\n  (the validator enforces it).\n- **Composite:** `overall = 100 · Σ wᵢ·scoreᵢ \u002F (4·Σ wᵢ)` with the critical cap (1 critical →\n  ≤60; ≥2 → poor). **Ship gate:** mean ≥3.0 AND no dimension \u003C2 AND zero critical errors.\n\ncolor-contrast carries the highest weight (**2.0**); any body text below 4.5:1 is a\ncritical that forces color-contrast to 0.\n\n---\n\n## Where this sits — the loop neighbors\n\n```\nG1 AUDIT ─ ui-audit (Track-A numbers + screenshots)   ──audit.json──▶  THIS (G3 CRITIQUE)\nG3 CRITIQUE ─ critique-panel  ◀ discuss edge          ──verdict──▶     G4 DIRECT (human decides)\n```\n\n**REQUIRED NEIGHBORS:** `ui-audit` (supplies `audit.json` + screenshots, shares the\nDESIGN-EVAL v1 rubric) · `design-studio` (the orchestrator that invokes this at G3) ·\n`frontend-design-direction` (G4, where the human picks among the enumerated directions).\n\n---\n\n## Terminal-state outputs\n\nEvery subcommand prints a marker on its last line (`AGENTS.md §4` anti-degeneration):\n- `SUCCESS:critique-panel-\u003Cpersona>-\u003Cid>` \u002F `FAILED:critique-panel-\u003Cpersona>-\u003Cid> \u003Creason>` (validate)\n- `SUCCESS:critique-panel-\u003Cid>` (synthesize)\n\n---\n\n## TDD\n\n```powershell\ncd ~\u002F.claude\u002Fskills\u002Fcritique-panel\nnpm test   # node --test tests\u002Fpanel_schema.test.mjs tests\u002Fcli.test.mjs\n```\n\nCovers: persona vocabulary + owned-dimension routing · the typed persona + synthesis\nschemas · the quota gate · the weighted composite + critical-error override\u002Fcap · and the\nCLI assemble\u002Fvalidate\u002Fsynthesize round-trip with terminal-state markers.\n","\u003C!-- origin: user (design-SOTA research, 2026-06-30); implements DESIGN-WORKFLOW.md G3 (NN\u002Fg Design Critique ritual + I Like\u002FI Wish\u002FWhat If + named multi-agent panel as TRIAGE), DESIGN-SKILLS-AND-EVAL.md §1.1 (one-persona-per-call reliability protocol), §1.2 (the four personas + quota), §2.1-§2.4 (0-4 scale, 9 dimensions\u002Fweights\u002Fcritical triggers, weighted composite + cap), §3 (defer DIRECTION to the human; enumerate don't pick). -->\n\n\n# critique-panel Skill\n\nThe **G3 CRITIQUE** gate of the TCR design-review loop — the *discuss edge*. Four named\nheuristic personas each reason over the **same** pre-computed Track-A numbers\n(`ui-audit` `audit.json`) + the screenshot, run the NN\u002Fg design-critique ritual, score\nthe nine **DESIGN-EVAL v1** dimensions 0–4, and emit a typed verdict. `synthesize` then\ndedupes + ranks them into one findings list with a weighted composite.\n\n**Skill location:** `~\u002F.claude\u002Fskills\u002Fcritique-panel\u002F`\n\n> The harness (`critique_panel.mjs`) is **pure plumbing** — it never calls a model and\n> never measures pixels. It routes evidence onto each persona's owned dimensions, builds\n> the per-persona brief, validates typed verdicts (incl. the quota gate), and aggregates.\n\n---\n\n## When to use\n\n- The **G3 CRITIQUE** gate after `ui-audit` (G1) has produced `audit.json` + screenshots.\n- You want a **multi-perspective**, severity-rated, typed critique (not \"looks good to me\").\n- You want the panel to **enumerate + rank** defects and propose directions for the human.\n\n**When NOT to use:**\n- Driving the app \u002F measuring pixels → that is `ui-audit` Track-A (this consumes its output).\n- Picking the winning redesign direction → the **human decides at G4** (`frontend-design-direction`).\n- The whole 8-gate loop end-to-end → `design-studio` (this is one leaf gate).\n\n---\n\n## The honesty note (cite it)\n\nThe panel is a **TRIAGE accelerator, NOT an oracle.** Multi-agent VLM critique has ~21%\nexpert overlap, ~24% false-positives, and is near-random at *picking* the better design\n(WiserUI-Bench). So the personas **enumerate, score, and rank** — the **human is the\ndecider at G4** (`DESIGN-SKILLS-AND-EVAL.md §3.2`).\n\n---\n\n## The four named personas\n\nEach persona stays in character and judges **only** its owned dimensions (one persona per\ncall kills the halo effect — `DESIGN-SKILLS-AND-EVAL.md §1.1`). Dimensions are\ndouble-covered on the high-weight \u002F critical-prone axes so the panel cross-checks.\n\n| Persona | Method | Owned dimensions |\n|---|---|---|\n| **user-advocate** | Nielsen's 10 usability heuristics | visual-hierarchy, empty-error-states, affordance, consistency |\n| **systems-consistency** | Refactoring UI + Material 3 \u002F HIG + Nielsen #4 | spacing-rhythm, type-scale, alignment, consistency |\n| **a11y** | WCAG 2.2 AA + WCAG 3.0 draft | color-contrast, affordance, motion |\n| **brand-aesthetic** | Refactoring UI colour\u002Fpersonality + Nielsen #8 | visual-hierarchy, color-contrast, type-scale, motion |\n\nThe full per-persona method + the 0–4 band, severity ladder, nine dimensions\u002Fweights, and\ncritical-error triggers live in **`panel_rubric.md`** (the deterministic lens, schema-aligned).\n\n---\n\n## Driving the panel (the harness)\n\n```powershell\n$CP = \"$HOME\u002F.claude\u002Fskills\u002Fcritique-panel\u002Fcritique_panel.mjs\"\n\n# 1) ASSEMBLE — read audit.json + screenshot(s) + rubric, route numbers onto each\n#    persona's owned dimensions, write FOUR persona briefs (.json + .md) to fill.\nnode $CP assemble --audit .\u002Fout\u002Faudit.json --screenshot a.png,b.png --scope \u002F,\u002Fsettings --out .\u002Fout\n\n# 2) (Claude, as each persona) fill panel-\u003Cpersona>.out.json — score 0-4 + I Like\u002FI Wish\u002FWhat If.\n\n# 3) VALIDATE — type-check a persona verdict incl. the quota gate (>=2 iLike + 1 iWish\u002Fscreen)\nnode $CP validate --input .\u002Fout\u002Fpanel-user-advocate.out.json\n#   -> SUCCESS:critique-panel-user-advocate-\u003Cid> \u002F FAILED:...\n\n# 4) SYNTHESIZE — merge verdicts -> ranked findings + weighted composite + protected wins\nnode $CP synthesize --inputs .\u002Fout\u002Fpanel-*.out.json --audit .\u002Fout\u002Faudit.json --out .\u002Fout\n#   -> SUCCESS:critique-panel-\u003Cid>\n\n# Dump the persona + synthesis JSON Schemas (keep them in sync)\nnode $CP schema --out .\u002F\n```\n\n**The contract:** the harness routes evidence and validates; **Claude reasons** as each\npersona over the numbers; `synthesize` computes the verdict. The personas reason over\nnumbers Track-A already measured — they never pixel-ground.\n\n---\n\n## Scoring (from `panel_rubric.md`)\n\n- **0–4 per owned dimension** + a `confidence`; `critical: true` only when that dimension's\n  critical-error trigger fired (forces its score to **0**).\n- **I Like** (wins to protect) \u002F **I Wish** (severity-rated defects + recommendation) \u002F\n  **What If** (alternative directions to enumerate — *not* pick).\n- **Quota (anti-tunnel-vision, even solo):** ≥2 `iLike` + 1 `iWish` per in-scope screen\n  (the validator enforces it).\n- **Composite:** `overall = 100 · Σ wᵢ·scoreᵢ \u002F (4·Σ wᵢ)` with the critical cap (1 critical →\n  ≤60; ≥2 → poor). **Ship gate:** mean ≥3.0 AND no dimension \u003C2 AND zero critical errors.\n\ncolor-contrast carries the highest weight (**2.0**); any body text below 4.5:1 is a\ncritical that forces color-contrast to 0.\n\n---\n\n## Where this sits — the loop neighbors\n\n```\nG1 AUDIT ─ ui-audit (Track-A numbers + screenshots)   ──audit.json──▶  THIS (G3 CRITIQUE)\nG3 CRITIQUE ─ critique-panel  ◀ discuss edge          ──verdict──▶     G4 DIRECT (human decides)\n```\n\n**REQUIRED NEIGHBORS:** `ui-audit` (supplies `audit.json` + screenshots, shares the\nDESIGN-EVAL v1 rubric) · `design-studio` (the orchestrator that invokes this at G3) ·\n`frontend-design-direction` (G4, where the human picks among the enumerated directions).\n\n---\n\n## Terminal-state outputs\n\nEvery subcommand prints a marker on its last line (`AGENTS.md §4` anti-degeneration):\n- `SUCCESS:critique-panel-\u003Cpersona>-\u003Cid>` \u002F `FAILED:critique-panel-\u003Cpersona>-\u003Cid> \u003Creason>` (validate)\n- `SUCCESS:critique-panel-\u003Cid>` (synthesize)\n\n---\n\n## TDD\n\n```powershell\ncd ~\u002F.claude\u002Fskills\u002Fcritique-panel\nnpm test   # node --test tests\u002Fpanel_schema.test.mjs tests\u002Fcli.test.mjs\n```\n\nCovers: persona vocabulary + owned-dimension routing · the typed persona + synthesis\nschemas · the quota gate · the weighted composite + critical-error override\u002Fcap · and the\nCLI assemble\u002Fvalidate\u002Fsynthesize round-trip with terminal-state markers.\n",{"name":5,"metadata":54,"description":7},{"origin":55},"user","main","a298c929b6cbd662e1fce54302ce520d5ceea1a5",".agents\u002Fskills\u002Fcritique-panel","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain","https:\u002F\u002Fgithub.com\u002FPrekzursil","2026-11-02 14:30:57",{"source":8,"skillName":5,"commands":63,"prompt":73,"zipAvailable":74},[64,67,70],{"tool":65,"command":66},"npx","npx skills add Prekzursil\u002Fagent-skills-toolchain --skill critique-panel",{"tool":68,"command":69},"bunx","bunx skills add Prekzursil\u002Fagent-skills-toolchain --skill critique-panel",{"tool":71,"command":72},"pnpm","pnpm dlx skills add Prekzursil\u002Fagent-skills-toolchain --skill critique-panel","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002FSKILL.md",true,[76,82,88,92,96,100,104,108,112],{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"critique_panel.mjs","SCRIPT","脚本",10422,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Fcritique_panel.mjs",{"path":83,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":86,"sourceHtmlUrl":87},"fixtures\u002Faudit.sample.json","REFERENCE","参考文档",2439,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Ffixtures\u002Faudit.sample.json",{"path":89,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":90,"sourceHtmlUrl":91},"package.json",264,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Fpackage.json",{"path":93,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":94,"sourceHtmlUrl":95},"panel_rubric.md",7492,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Fpanel_rubric.md",{"path":97,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":98,"sourceHtmlUrl":99},"panel_schema.mjs",33700,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Fpanel_schema.mjs",{"path":101,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":102,"sourceHtmlUrl":103},"persona.schema.json",2977,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Fpersona.schema.json",{"path":105,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":106,"sourceHtmlUrl":107},"synthesis.schema.json",3865,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Fsynthesis.schema.json",{"path":109,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":110,"sourceHtmlUrl":111},"tests\u002Fcli.test.mjs",2820,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Ftests\u002Fcli.test.mjs",{"path":113,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":114,"sourceHtmlUrl":115},"tests\u002Fpanel_schema.test.mjs",9554,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcritique-panel\u002Ftests\u002Fpanel_schema.test.mjs",[117,138,167,192,206,221],{"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":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":49},"01M6DF7D7478R55P46REYH4QAC","clarify-requirements","Resolve genuine pre-execution ambiguity by scanning repo context, generating 1-6 grounded options, and asking 1-3 focused questions via AskUserQuestion. NOT for mid-task use (injection risk); NOT when the goal is already unambiguous; use to disambiguate vague prompts before any file edits.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fworkspace-hooks\u002Fskills\u002Fclarify-requirements\u002FSKILL.md",[],[124,125,126,127,128,129,130,131,132,37,133,134,135,136,137],"clarify requirements","clarify","requirements","risk","options","task","prompts","goal","scanning","context","repo","injection","execution","file",{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":147,"publishedAt":49},"01M6DF79CA4BHJS66J8EZDBPWJ","openmm-md","Run FREE, open-source, Windows-native GPU molecular dynamics via the openmm-md MCP (OpenMM + CUDA on the local NVIDIA GPU). Use for protein \u002F protein-ligand MD when you want a free stack or are on Windows without a Schrodinger license. Covers the full workflow (prepare, solvate, minimize, equilibrate, production, analyze, mmgbsa) AND an honest decision guide vs Schrodinger Desmond, including how to run Desmond legitimately on Linux or a cloud GPU with a real license. This is a deliberately LESSER free alternative and does NOT replace the schrodinger MCP tools. Do NOT use it to pick a free tool when a validated FEP+, WaterMap, IFD-MD, or Prime MM-GBSA result is actually required.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fmcp\u002Fopenmm-md-mcp\u002Fskill\u002Fopenmm-md\u002FSKILL.md",[144],{"categoryNo":145,"categoryName":146,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[148,149,150,151,152,153,154,155,156,157,158,159,160,161,37,162,163,164,165,166],"openmm md","how","decision","pick","linux","analyze","tool","stack","open","NVIDIA","gpu","license","workflow","windows","free","molecular dynamics","local","do","MCP",{"skillNo":168,"name":169,"nameZh":6,"description":170,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":171,"categories":172,"keywords":179,"publishedAt":49},"01M6DF76C1MG85VEWP8VZCBNMN","visual-review","Take screenshots of web pages and UI using Playwright for visual review and iteration","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fvisual-review\u002FSKILL.md",[173,176],{"categoryNo":174,"categoryName":175,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000003","演示文稿",{"categoryNo":177,"categoryName":178,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量",[48,180,181,182,183,184,185,186,187,188,149,189,45,190,191],"visual","visual review","pages","web","false","iteration","take","check","slides","ui","playwright","screenshots",{"skillNo":193,"name":194,"nameZh":6,"description":195,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":196,"categories":197,"keywords":198,"publishedAt":49},"01M6DF73AQVCBZKEN0YXEX9FXP","status","Diagnostic status report — shows metaswarm installation state, project setup, and potential issues","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstatus\u002FSKILL.md",[],[194,199,200,201,202,203,204,205],"setup","report","project","installation","project setup","issues","state",{"skillNo":207,"name":208,"nameZh":6,"description":209,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":210,"categories":211,"keywords":213,"publishedAt":49},"01M6DF6X6C2XB0P6GX76P7TTYV","start","Use when starting work on any task, when the user mentions metaswarm, or when the user wants to begin tracked development work","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstart\u002FSKILL.md",[212],{"categoryNo":145,"categoryName":146,"taxonomyType":21,"taxonomyTypeDesc":22},[208,214,129,215,216,217,37,55,218,219,220],"start issue","issue","ready","work on issue","work","work on","development",{"skillNo":222,"name":199,"nameZh":6,"description":223,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":224,"categories":225,"keywords":226,"publishedAt":49},"01M6DF6QMJ9AJKFKHHKRT005PQ","Interactive project setup — detects your project, configures metaswarm, writes project-local files","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fsetup\u002FSKILL.md",[],[199,201,164,227,203,228],"files","interactive"]