[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ABAS4F1SYSE2CJTR287KBB":3,"skill-same:01M7ABAS4F1SYSE2CJTR287KBB":50},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":14,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":49},"01M7ABAS4F1SYSE2CJTR287KBB","review-skill",null,"Reviews, audits, and scores skill files (SKILL.md) for quality, completeness, frontmatter correctness, discoverability, and anti-patterns. Use when asked to review, audit, check, or improve a skill file, or when a user says \"is this skill good?\", \"check my skill\", or pastes a SKILL.md for feedback. Also use proactively after writing a new skill — self-apply as a final quality gate. Don't use for reviewing general Markdown documentation, README files, or code files that are not skill definitions.","ism200\u002Fpygta-workspace","ism200","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F18715155?v=4","01M7ABA5J5HGGD8DGKM7Y3GNGN","Ivo van Stokkum","01M7ABA5JSA8MNGGG64QRBDTXT",0,"MIT","https:\u002F\u002Fgithub.com\u002Fism200\u002Fpygta-workspace\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Freview-skill\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: review-skill\ndescription: >\n  Reviews, audits, and scores skill files (SKILL.md) for quality,\n  completeness, frontmatter correctness, discoverability, and\n  anti-patterns. Use when asked to review, audit, check, or improve\n  a skill file, or when a user says \"is this skill good?\", \"check my\n  skill\", or pastes a SKILL.md for feedback. Also use proactively\n  after writing a new skill — self-apply as a final quality gate.\n  Don't use for reviewing general Markdown documentation, README\n  files, or code files that are not skill definitions.\nargument-hint: \"Optional: skill name or path. Omit to audit all discovered skills.\"\nmetadata:\n  version: \"1.0\"\n---\n\n# Review Skill\n\nAudit `SKILL.md` files against the [Agent Skills specification](https:\u002F\u002Fagentskills.io\u002Fspecification) and produce a scored Markdown report.\n\n## Guardrails\n\n- Only review `SKILL.md` files — not general Markdown, READMEs, or code.\n- Collect **all** discovered skills before reviewing any.\n- This skill can review itself as a smoke test.\n\n## Step 1: Discovery\n\nIf the user provided a specific skill name or path, review only that\nskill and skip the search below.\n\nOtherwise search every location:\n\n| Priority | Location             | Scope                                 |\n| -------- | -------------------- | ------------------------------------- |\n| 1        | `skills\u002F`            | repo-level reference skills           |\n| 2        | `.github\u002Fskills\u002F`    | GitHub Copilot project skills         |\n| 3        | `.agents\u002Fskills\u002F`    | standardized agent skills (preferred) |\n| 4        | `.claude\u002Fskills\u002F`    | Claude agent skills                   |\n| 5        | `~\u002F.copilot\u002Fskills\u002F` | user-level Copilot skills             |\n| 6        | `~\u002F.agents\u002Fskills\u002F`  | user-level agent skills (preferred)   |\n| 7        | `~\u002F.claude\u002Fskills\u002F`  | user-level Claude skills              |\n\nA skill is a **subfolder** containing `SKILL.md` (preferred) or\n`SKILLS.md` (non-standard — flag it). Loose `.md` files are not skills.\n\nIf the same skill name appears in multiple locations, report the\nconflict and note which location wins at runtime (higher priority\nnumber = lower precedence).\n\n## Step 2: Review each skill\n\nApply every check below. Assign severity: **Error** (blocks correct\nfunction), **Warning** (degrades quality or discoverability),\n**Info** (style suggestion).\n\n### File and folder\n\n| Check                    | Pass condition              | Severity if failed |\n| ------------------------ | --------------------------- | ------------------ |\n| Skill in named subfolder | Not loose in skills root    | Error              |\n| File named `SKILL.md`    | Exact match, case-sensitive | Warning            |\n| Folder name format       | `[a-z0-9-]+` only           | Warning            |\n\n### Frontmatter\n\nValidate against the [Agent Skills specification](https:\u002F\u002Fagentskills.io\u002Fspecification).\nSkills in agent-loaded locations (`.agents\u002F`, `.github\u002F`, `.copilot\u002F`,\n`.claude\u002F`) **must** have frontmatter. Skills in a root `skills\u002F`\nfolder intended as human-readable reference may omit it — note but\ndon't error.\n\n**Errors** (block correct function):\n- Frontmatter block missing (`---` delimiters not found) — Error for agent-loaded, Info for reference skills\n- `name` does not match folder name exactly\n- `description` missing or empty\n\n**Warnings** (degrade quality):\n- `description` under 20 chars\n- `description` lacks searchable trigger phrases\n- Unknown frontmatter fields (spec allows: `license`, `compatibility`, `metadata`, `allowed-tools`)\n\n**Info:**\n- No negative trigger (\"Don't use when...\") in description\n\n### Body content\n\n**Errors:**\n- No numbered, actionable procedure steps\n- `.\u002Fpath` references that don't resolve to a file\n\n**Warnings:**\n- Over 500 lines (recommend splitting to `references\u002F`)\n- Absolute paths (`\u002Fhome\u002F...`) instead of `.\u002F`\n- Time-sensitive logic (hard-coded dates or version cutoffs)\n- Future-tense only — has promises but no current steps\n\n**Info:**\n- Body opens with a \"Purpose and contract\" section instead of a brief orienting sentence or guardrails\n- No guardrails section (if applicable)\n- No output contract (if the skill produces artifacts)\n\n### Quality and discoverability\n\n- Would this description cause the agent to select the skill for the right tasks?\n- Are procedures specific enough to produce consistent output across runs?\n- Is bulk content pushed to `references\u002F` rather than inline?\n- Does the argument-hint (if declared) have corresponding branch logic in the procedure?\n\n## Step 3: Anti-patterns\n\nFlag if present — these are the most common skill authoring mistakes:\n\n| Anti-Pattern           | Explanation                                       |\n| ---------------------- | ------------------------------------------------- |\n| Vague description      | \"A helpful skill\" — agent cannot route to it      |\n| No negative trigger    | Adjacent tasks will false-positive                |\n| Monolithic body        | Everything inline; no `references\u002F` offload       |\n| Missing procedure      | Preamble only, no actionable steps                |\n| Unused declared fields | `argument-hint` in frontmatter, no branch in body |\n\n## Step 4: Output\n\nProduce a Markdown report:\n\n```markdown\n# Skill Review Report\n\n## Summary\n\n- Skills found: N\n- Locations searched: [list]\n- Issues: N errors, N warnings, N info\n\n## Skills Reviewed\n\n### `\u003Cfolder-name>`\n\n**Location**: `\u003Crelative path to SKILL.md>`\n**Status**: ✅ Pass \u002F ⚠️ Warnings \u002F ❌ Errors\n\n| Check               | Severity | Result | Note                                |\n| ------------------- | -------- | ------ | ----------------------------------- |\n| Folder naming       | —        | ✅     |                                     |\n| File name           | Warning  | ⚠️     | Uses SKILLS.md                      |\n| Frontmatter present | Error    | ❌     | No --- block                        |\n| name matches folder | —        | N\u002FA    | Cannot check (no frontmatter)       |\n| description quality | Warning  | ⚠️     | 12 chars, too short                 |\n| Negative trigger    | Info     | ℹ️     | Consider adding \"Don't use when...\" |\n| Procedure present   | —        | ✅     |                                     |\n| File length         | —        | ✅     | 94 lines                            |\n\n**Recommendations** (ordered by severity):\n\n1. ❌ Add YAML frontmatter with `name` and `description`\n2. ⚠️ Rename `SKILLS.md` → `SKILL.md`\n3. ⚠️ Expand description to include trigger phrases\n4. ℹ️ Add a negative trigger to reduce false positives\n\n---\n```\n\nIf only one skill was reviewed (scoped run), omit the Summary section\nand go straight to the skill block.\n","# Review Skill\n\nAudit `SKILL.md` files against the [Agent Skills specification](https:\u002F\u002Fagentskills.io\u002Fspecification) and produce a scored Markdown report.\n\n## Guardrails\n\n- Only review `SKILL.md` files — not general Markdown, READMEs, or code.\n- Collect **all** discovered skills before reviewing any.\n- This skill can review itself as a smoke test.\n\n## Step 1: Discovery\n\nIf the user provided a specific skill name or path, review only that\nskill and skip the search below.\n\nOtherwise search every location:\n\n| Priority | Location             | Scope                                 |\n| -------- | -------------------- | ------------------------------------- |\n| 1        | `skills\u002F`            | repo-level reference skills           |\n| 2        | `.github\u002Fskills\u002F`    | GitHub Copilot project skills         |\n| 3        | `.agents\u002Fskills\u002F`    | standardized agent skills (preferred) |\n| 4        | `.claude\u002Fskills\u002F`    | Claude agent skills                   |\n| 5        | `~\u002F.copilot\u002Fskills\u002F` | user-level Copilot skills             |\n| 6        | `~\u002F.agents\u002Fskills\u002F`  | user-level agent skills (preferred)   |\n| 7        | `~\u002F.claude\u002Fskills\u002F`  | user-level Claude skills              |\n\nA skill is a **subfolder** containing `SKILL.md` (preferred) or\n`SKILLS.md` (non-standard — flag it). Loose `.md` files are not skills.\n\nIf the same skill name appears in multiple locations, report the\nconflict and note which location wins at runtime (higher priority\nnumber = lower precedence).\n\n## Step 2: Review each skill\n\nApply every check below. Assign severity: **Error** (blocks correct\nfunction), **Warning** (degrades quality or discoverability),\n**Info** (style suggestion).\n\n### File and folder\n\n| Check                    | Pass condition              | Severity if failed |\n| ------------------------ | --------------------------- | ------------------ |\n| Skill in named subfolder | Not loose in skills root    | Error              |\n| File named `SKILL.md`    | Exact match, case-sensitive | Warning            |\n| Folder name format       | `[a-z0-9-]+` only           | Warning            |\n\n### Frontmatter\n\nValidate against the [Agent Skills specification](https:\u002F\u002Fagentskills.io\u002Fspecification).\nSkills in agent-loaded locations (`.agents\u002F`, `.github\u002F`, `.copilot\u002F`,\n`.claude\u002F`) **must** have frontmatter. Skills in a root `skills\u002F`\nfolder intended as human-readable reference may omit it — note but\ndon't error.\n\n**Errors** (block correct function):\n- Frontmatter block missing (`---` delimiters not found) — Error for agent-loaded, Info for reference skills\n- `name` does not match folder name exactly\n- `description` missing or empty\n\n**Warnings** (degrade quality):\n- `description` under 20 chars\n- `description` lacks searchable trigger phrases\n- Unknown frontmatter fields (spec allows: `license`, `compatibility`, `metadata`, `allowed-tools`)\n\n**Info:**\n- No negative trigger (\"Don't use when...\") in description\n\n### Body content\n\n**Errors:**\n- No numbered, actionable procedure steps\n- `.\u002Fpath` references that don't resolve to a file\n\n**Warnings:**\n- Over 500 lines (recommend splitting to `references\u002F`)\n- Absolute paths (`\u002Fhome\u002F...`) instead of `.\u002F`\n- Time-sensitive logic (hard-coded dates or version cutoffs)\n- Future-tense only — has promises but no current steps\n\n**Info:**\n- Body opens with a \"Purpose and contract\" section instead of a brief orienting sentence or guardrails\n- No guardrails section (if applicable)\n- No output contract (if the skill produces artifacts)\n\n### Quality and discoverability\n\n- Would this description cause the agent to select the skill for the right tasks?\n- Are procedures specific enough to produce consistent output across runs?\n- Is bulk content pushed to `references\u002F` rather than inline?\n- Does the argument-hint (if declared) have corresponding branch logic in the procedure?\n\n## Step 3: Anti-patterns\n\nFlag if present — these are the most common skill authoring mistakes:\n\n| Anti-Pattern           | Explanation                                       |\n| ---------------------- | ------------------------------------------------- |\n| Vague description      | \"A helpful skill\" — agent cannot route to it      |\n| No negative trigger    | Adjacent tasks will false-positive                |\n| Monolithic body        | Everything inline; no `references\u002F` offload       |\n| Missing procedure      | Preamble only, no actionable steps                |\n| Unused declared fields | `argument-hint` in frontmatter, no branch in body |\n\n## Step 4: Output\n\nProduce a Markdown report:\n\n```markdown\n# Skill Review Report\n\n## Summary\n\n- Skills found: N\n- Locations searched: [list]\n- Issues: N errors, N warnings, N info\n\n## Skills Reviewed\n\n### `\u003Cfolder-name>`\n\n**Location**: `\u003Crelative path to SKILL.md>`\n**Status**: ✅ Pass \u002F ⚠️ Warnings \u002F ❌ Errors\n\n| Check               | Severity | Result | Note                                |\n| ------------------- | -------- | ------ | ----------------------------------- |\n| Folder naming       | —        | ✅     |                                     |\n| File name           | Warning  | ⚠️     | Uses SKILLS.md                      |\n| Frontmatter present | Error    | ❌     | No --- block                        |\n| name matches folder | —        | N\u002FA    | Cannot check (no frontmatter)       |\n| description quality | Warning  | ⚠️     | 12 chars, too short                 |\n| Negative trigger    | Info     | ℹ️     | Consider adding \"Don't use when...\" |\n| Procedure present   | —        | ✅     |                                     |\n| File length         | —        | ✅     | 94 lines                            |\n\n**Recommendations** (ordered by severity):\n\n1. ❌ Add YAML frontmatter with `name` and `description`\n2. ⚠️ Rename `SKILLS.md` → `SKILL.md`\n3. ⚠️ Expand description to include trigger phrases\n4. ℹ️ Add a negative trigger to reduce false positives\n\n---\n```\n\nIf only one skill was reviewed (scoped run), omit the Summary section\nand go straight to the skill block.\n",{"name":5,"metadata":24,"description":26,"argument-hint":27},{"version":25},"1.0","Reviews, audits, and scores skill files (SKILL.md) for quality, completeness, frontmatter correctness, discoverability, and anti-patterns. Use when asked to review, audit, check, or improve a skill file, or when a user says \"is this skill good?\", \"check my skill\", or pastes a SKILL.md for feedback. Also use proactively after writing a new skill — self-apply as a final quality gate. Don't use for reviewing general Markdown documentation, README files, or code files that are not skill definitions.\n","Optional: skill name or path. Omit to audit all discovered skills.","main","1a42aea92d71c3a655516d0466d1a433247b3024",".agents\u002Fskills\u002Freview-skill","https:\u002F\u002Fgithub.com\u002Fism200\u002Fpygta-workspace","https:\u002F\u002Fgithub.com\u002Fism200","http:\u002F\u002Fwww.few.vu.nl\u002F~ivo\u002F","Associate Professor Biophysics at Vrije Universiteit Amsterdam\r\n\r\nSpecialized in global and target analysis of time-resolved spectra","2026-07-11 01:13:08",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add ism200\u002Fpygta-workspace --skill review-skill",{"tool":42,"command":43},"bunx","bunx skills add ism200\u002Fpygta-workspace --skill review-skill",{"tool":45,"command":46},"pnpm","pnpm dlx skills add ism200\u002Fpygta-workspace --skill review-skill","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fism200\u002Fpygta-workspace\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Freview-skill\u002FSKILL.md",true,[],[51,58],{"skillNo":52,"name":53,"nameZh":6,"description":54,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":19},"01M7ABAP4ABM7A0BK2YFT0CTWH","pygta-validate","Run validation across the workspace root and configured external repos with changes.","https:\u002F\u002Fgithub.com\u002Fism200\u002Fpygta-workspace\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpygta-validate\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M7ABAKGF74BEZFHDN13FC89Z","pygta-project","Scaffold a new pyglotaran analysis project with standard folder structure and starter notebook. Don't use when modifying an existing project or changing the scaffold script itself","https:\u002F\u002Fgithub.com\u002Fism200\u002Fpygta-workspace\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpygta-project\u002FSKILL.md",[],[]]