[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7C8YV6AAFSCTS39C2MBKTTV":3,"skill-same:01M7C8YV6AAFSCTS39C2MBKTTV":52},{"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":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7C8YV6AAFSCTS39C2MBKTTV","write-skills",null,"Create or revise agent skills. Use when adding a new skill file, renaming a skill, simplifying an existing skill, improving trigger descriptions, or deciding what belongs in a skill versus references, scripts, assets, or ordinary docs.","transitive-bullshit\u002Fskills","transitive-bullshit","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F552829?v=4","01M7C8KS4F4572TZFKR12N2A3W","Travis Fischer","01M7C8KS56YA2KDQHDNXEGS23P",4,"MIT","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwrite-skills\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwrite-skills\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: write-skills\ndescription: Create or revise agent skills. Use when adding a new skill file, renaming a skill, simplifying an existing skill, improving trigger descriptions, or deciding what belongs in a skill versus references, scripts, assets, or ordinary docs.\n---\n\n# Write Skills\n\nA skill is not documentation. It is compressed operational memory for an\nagent that already knows how to code and reason. Its job is\n**predictability** — the agent taking the same *process* every run, not\nproducing the same output. Put only the context that changes what the agent\nwill do.\n\n## First Principles\n\n1. **Trigger from the description.** The frontmatter `description` is the\n   only part read before the skill loads. Say what the skill does and the\n   concrete situations that should trigger it. Do not hide trigger rules in\n   the body.\n\n2. **Spend tokens like they are scarce.** Assume the agent is already good at\n   general reasoning. State the direction and let smart models figure out\n   implementation details. A wrapper skill can be two sentences: name the\n   capability to use and the skill to follow. Add mechanics only when a\n   demonstrated failure requires them.\n\n3. **Write procedures, not essays.** Prefer imperative rules, decision\n   points, and small examples. A good skill changes behavior in the next\n   turn; it does not merely explain the topic. And match the procedure to\n   the deliverable's shape: a catalog of techniques with \"pick what fits\"\n   produces tool-picking, not a flow — if the job is a shaped interaction\n   (an interview, a staged walk), the workflow must *be* that shape, with\n   the techniques demoted to steps inside it.\n\n4. **Use progressive disclosure.** Keep the main skill file short. Push long\n   schemas, examples, provider docs, or variant-specific guidance out into\n   linked `references\u002F` files, reached by a pointer that fires only when\n   needed. Put repeatable fragile operations in `scripts\u002F`. Put reusable\n   output material in `assets\u002F`.\n\n5. **Validate by use.** A skill is good when a fresh agent applies it\n   correctly on a realistic task — run it blind, and on the weakest model\n   that will run it; a skill that only drives correct behavior on the\n   frontier model that authored it is too fragile to ship. After editing,\n   read it as if you had no conversation history and remove anything that\n   would not affect action.\n\n6. **Examples document the PROBLEM, not the solution.** An example earns its\n   tokens by teaching the agent to *recognise a recurring problem* — the smell,\n   the symptom, how you knew it was wrong. That is durable. The fix you happened\n   to apply is not: code changes, and a baked-in solution goes stale, or worse\n   prescribes a move that won't fit next time. Write the failure mode and its\n   tell; let the agent derive the fix fresh against the current code. \"A\n   directional question gated on a centroid distance read wrong from every\n   bearing\" teaches; \"so we keyed it on FRONT_ARC\" rots. When in doubt, state\n   what was broken and how you spotted it, and stop there.\n\n## Leading Words\n\nA **leading word** is a compact concept already in the model's pretraining\nthat the agent thinks with while running the skill (e.g. *fog of war*,\n*tracer bullets*, *red*, *tight*). One word recruits priors the model\nalready holds and anchors a whole region of behavior in the fewest tokens.\n\n- It anchors **execution** in the body (same word → same behavior every\n  time) and **invocation** in the description (when that word also lives in\n  the user's prompts, docs, and code, the skill fires more reliably).\n- Hunt for restatements that a leading word retires. \"fast, deterministic,\n  low-overhead\" → a *tight* loop. \"a loop you believe in\" → the loop goes\n  *red*. You win twice: fewer tokens and a sharper hook.\n- A weak leading word is a no-op (`be thorough` when the agent already is).\n  The fix is a stronger word (*relentless*), not more sentences.\n\n## Invocation\n\nChoose how the skill is reached; each choice spends a different cost.\n\n- **Model-invoked** (default): keep a `description` so the agent can fire it\n  on its own and other skills can reach it. Costs **context load** — the\n  description sits in the window every turn. Write rich trigger phrasing.\n- **User-invoked**: when explicitly requested, set\n  `policy.allow_implicit_invocation: false` in `agents\u002Fopenai.yaml` for Codex.\n  Preserve unrelated existing metadata. Other clients may also need\n  `disable-model-invocation: true` in frontmatter; that field alone does not\n  configure Codex. The user can still invoke the skill by name. Verify the\n  actual host behavior rather than assuming zero context cost.\n\nPreserve the current invocation policy. New skills remain model-invoked unless\nthe user requests explicit-only invocation. When user-invoked skills pile up\npast memory, add a **router skill** that names the others and when to reach\nfor each.\n\n## Description\n\nThe description does two jobs: state what the skill is, and list the\ndistinct situations (**branches**) that trigger it.\n\n- Front-load the leading word — invocation work happens here.\n- One trigger per branch. Synonyms that rename one branch are duplication;\n  collapse them and keep only genuinely distinct branches.\n- Cut identity already in the body. Keep triggers plus any \"when another\n  skill needs…\" reach clause.\n\n## Information Hierarchy\n\nA skill is **steps** and **reference**, mixed freely. Rank each piece by how\nimmediately the agent needs it:\n\n1. **In-skill step** — an ordered action in `SKILL.md`. Each ends on a\n   **completion criterion**: make it *checkable* (can the agent tell done\n   from not-done?) and, where it matters, *exhaustive* (\"every modified\n   model accounted for\", not \"produce a change list\"). A vague criterion\n   invites premature completion.\n2. **In-skill reference** — a definition or rule consulted on demand. A flat\n   peer-set of rules is fine, not a smell.\n3. **External reference** — pushed out of `SKILL.md` into a linked file,\n   loaded only when its pointer fires.\n\nKeep a concept's definition, rules, and caveats under one heading\n(**co-location**) so reading one part brings its neighbors. Push too little\ndown and the top bloats; push too much down and you hide what the agent needs.\n\n## When to Split\n\nEach cut spends a cost, so split only when it earns it:\n\n- **By invocation** — split off a model-invoked skill when a distinct\n  leading word should trigger it alone, or another skill must reach it. You\n  pay context load for the new always-loaded description.\n- **By sequence** — split a run of steps when the steps still ahead tempt\n  the agent to rush the one in front of it. Hiding later steps forces more\n  work on the current one. The tell is **stage compression**: several steps'\n  work lands in one message, or the agent narrates a later step as complete\n  without ever having opened it. A \"when you enter a step, read its file\"\n  pointer in the main skill fires reliably, even on weaker models.\n\n## Failure Modes\n\nDiagnose a misbehaving skill against these:\n\n- **Premature completion** — ending a step before it's done. Fix the\n  completion criterion first (cheap); only split to hide later steps if the\n  criterion is irreducibly fuzzy *and* you see the rush. When the skill's\n  job ends in a handover artifact, name it as the *only* skill-level\n  done-condition — otherwise runs end at whichever intermediate artifact\n  feels finished.\n- **Embargo** — an ordered workflow over-obeyed: the agent withholds a\n  finding made early to honor a later step's choreography, so the user\n  decides something while the agent sits on information that bears on it.\n  Any skill that sequences steps needs the escape valve stated: order\n  governs presentation, never disclosure.\n- **Lucky pass** — a validation run that succeeds only because the user or\n  world volunteered a critical input unprompted. The outcome was right but\n  the process didn't produce it; encode the eliciting probe as an explicit\n  step instead of banking on the luck recurring.\n- **Duplication** — the same meaning in two places. Keep a **single source\n  of truth**.\n- **Sediment** — stale layers that accumulate because adding feels safe.\n  Prune deliberately.\n- **War story** — a lesson written as the play-by-play of the change that\n  taught it: function names, tuned values, one bug's trajectory. Those\n  specifics date fast and bury the transferable rule. State the principle and\n  the smell to watch for; let the codebase hold the mechanics. One concrete\n  touchstone grounds it; a paragraph of them drowns it.\n- **Implementation index** — a skill that points at today's source files, line\n  numbers, current literals, or exact internal functions when its job is really\n  to teach judgment. Those locators rot and make agents chase old mechanics.\n  Keep durable principles, symptoms, acceptance criteria, and reference assets\n  in the skill; put task-specific implementation notes in the active spec. If\n  code location matters, tell the agent to find the current owner in the\n  codebase.\n- **Sprawl** — too long even when every line is live. Cure with the ladder:\n  disclose reference behind pointers, split by branch or sequence.\n- **No-op** — a line the model already obeys by default. Test each sentence\n  in isolation; when it fails, delete the whole sentence, don't trim words.\n\n## Shape\n\nUse this structure unless there is a strong reason not to:\n\n```markdown\n---\nname: short-verb-phrase\ndescription: What this does. Use when ...\n---\n\n# Skill Title\n\nOne short paragraph defining the job.\n\n## Workflow\n\n1. Do the first load-bearing thing.\n2. Make the key decision.\n3. Produce or verify the artifact.\n\n## Rules\n\n- Keep the constraints that prevent common mistakes.\n- Link only the references that should be loaded conditionally.\n```\n\n## Edit Pass\n\nWhen creating or revising a skill:\n\n- Prefer **action + plural noun** in lowercase hyphen-case: `audit-tests`,\n  `write-skills`, `compare-screenshots`. Keep the folder and frontmatter name\n  identical. Use a natural exception when pluralizing would distort the meaning;\n  this is the default for new names, not a mandate to rename existing skills.\n- Make the description specific enough to trigger without the body.\n- Remove any \"when to use\" section from the body.\n- Remove stale history, placeholders, and setup notes. Preserve attribution,\n  provenance records, and required license notices.\n- Remove file paths, line numbers, current constants, and implementation knobs\n  unless the skill is explicitly a code-navigation runbook. Prefer durable\n  principles plus a directive to inspect the current code.\n- Prefer one strong rule over several overlapping bullets.\n- Refactor restatements into a leading word where one fits.\n- Keep examples tiny and realistic.\n- Add no README, changelog, or auxiliary docs unless they are actual\n  references the skill tells the agent when to read.\n- Run the skill validator when available.\n\n## Done\n\nThe skill is done when its metadata triggers correctly, its body is short\nenough to read in one pass, and a fresh agent can follow it without asking\nwhy the skill exists.\n","# Write Skills\n\nA skill is not documentation. It is compressed operational memory for an\nagent that already knows how to code and reason. Its job is\n**predictability** — the agent taking the same *process* every run, not\nproducing the same output. Put only the context that changes what the agent\nwill do.\n\n## First Principles\n\n1. **Trigger from the description.** The frontmatter `description` is the\n   only part read before the skill loads. Say what the skill does and the\n   concrete situations that should trigger it. Do not hide trigger rules in\n   the body.\n\n2. **Spend tokens like they are scarce.** Assume the agent is already good at\n   general reasoning. State the direction and let smart models figure out\n   implementation details. A wrapper skill can be two sentences: name the\n   capability to use and the skill to follow. Add mechanics only when a\n   demonstrated failure requires them.\n\n3. **Write procedures, not essays.** Prefer imperative rules, decision\n   points, and small examples. A good skill changes behavior in the next\n   turn; it does not merely explain the topic. And match the procedure to\n   the deliverable's shape: a catalog of techniques with \"pick what fits\"\n   produces tool-picking, not a flow — if the job is a shaped interaction\n   (an interview, a staged walk), the workflow must *be* that shape, with\n   the techniques demoted to steps inside it.\n\n4. **Use progressive disclosure.** Keep the main skill file short. Push long\n   schemas, examples, provider docs, or variant-specific guidance out into\n   linked `references\u002F` files, reached by a pointer that fires only when\n   needed. Put repeatable fragile operations in `scripts\u002F`. Put reusable\n   output material in `assets\u002F`.\n\n5. **Validate by use.** A skill is good when a fresh agent applies it\n   correctly on a realistic task — run it blind, and on the weakest model\n   that will run it; a skill that only drives correct behavior on the\n   frontier model that authored it is too fragile to ship. After editing,\n   read it as if you had no conversation history and remove anything that\n   would not affect action.\n\n6. **Examples document the PROBLEM, not the solution.** An example earns its\n   tokens by teaching the agent to *recognise a recurring problem* — the smell,\n   the symptom, how you knew it was wrong. That is durable. The fix you happened\n   to apply is not: code changes, and a baked-in solution goes stale, or worse\n   prescribes a move that won't fit next time. Write the failure mode and its\n   tell; let the agent derive the fix fresh against the current code. \"A\n   directional question gated on a centroid distance read wrong from every\n   bearing\" teaches; \"so we keyed it on FRONT_ARC\" rots. When in doubt, state\n   what was broken and how you spotted it, and stop there.\n\n## Leading Words\n\nA **leading word** is a compact concept already in the model's pretraining\nthat the agent thinks with while running the skill (e.g. *fog of war*,\n*tracer bullets*, *red*, *tight*). One word recruits priors the model\nalready holds and anchors a whole region of behavior in the fewest tokens.\n\n- It anchors **execution** in the body (same word → same behavior every\n  time) and **invocation** in the description (when that word also lives in\n  the user's prompts, docs, and code, the skill fires more reliably).\n- Hunt for restatements that a leading word retires. \"fast, deterministic,\n  low-overhead\" → a *tight* loop. \"a loop you believe in\" → the loop goes\n  *red*. You win twice: fewer tokens and a sharper hook.\n- A weak leading word is a no-op (`be thorough` when the agent already is).\n  The fix is a stronger word (*relentless*), not more sentences.\n\n## Invocation\n\nChoose how the skill is reached; each choice spends a different cost.\n\n- **Model-invoked** (default): keep a `description` so the agent can fire it\n  on its own and other skills can reach it. Costs **context load** — the\n  description sits in the window every turn. Write rich trigger phrasing.\n- **User-invoked**: when explicitly requested, set\n  `policy.allow_implicit_invocation: false` in `agents\u002Fopenai.yaml` for Codex.\n  Preserve unrelated existing metadata. Other clients may also need\n  `disable-model-invocation: true` in frontmatter; that field alone does not\n  configure Codex. The user can still invoke the skill by name. Verify the\n  actual host behavior rather than assuming zero context cost.\n\nPreserve the current invocation policy. New skills remain model-invoked unless\nthe user requests explicit-only invocation. When user-invoked skills pile up\npast memory, add a **router skill** that names the others and when to reach\nfor each.\n\n## Description\n\nThe description does two jobs: state what the skill is, and list the\ndistinct situations (**branches**) that trigger it.\n\n- Front-load the leading word — invocation work happens here.\n- One trigger per branch. Synonyms that rename one branch are duplication;\n  collapse them and keep only genuinely distinct branches.\n- Cut identity already in the body. Keep triggers plus any \"when another\n  skill needs…\" reach clause.\n\n## Information Hierarchy\n\nA skill is **steps** and **reference**, mixed freely. Rank each piece by how\nimmediately the agent needs it:\n\n1. **In-skill step** — an ordered action in `SKILL.md`. Each ends on a\n   **completion criterion**: make it *checkable* (can the agent tell done\n   from not-done?) and, where it matters, *exhaustive* (\"every modified\n   model accounted for\", not \"produce a change list\"). A vague criterion\n   invites premature completion.\n2. **In-skill reference** — a definition or rule consulted on demand. A flat\n   peer-set of rules is fine, not a smell.\n3. **External reference** — pushed out of `SKILL.md` into a linked file,\n   loaded only when its pointer fires.\n\nKeep a concept's definition, rules, and caveats under one heading\n(**co-location**) so reading one part brings its neighbors. Push too little\ndown and the top bloats; push too much down and you hide what the agent needs.\n\n## When to Split\n\nEach cut spends a cost, so split only when it earns it:\n\n- **By invocation** — split off a model-invoked skill when a distinct\n  leading word should trigger it alone, or another skill must reach it. You\n  pay context load for the new always-loaded description.\n- **By sequence** — split a run of steps when the steps still ahead tempt\n  the agent to rush the one in front of it. Hiding later steps forces more\n  work on the current one. The tell is **stage compression**: several steps'\n  work lands in one message, or the agent narrates a later step as complete\n  without ever having opened it. A \"when you enter a step, read its file\"\n  pointer in the main skill fires reliably, even on weaker models.\n\n## Failure Modes\n\nDiagnose a misbehaving skill against these:\n\n- **Premature completion** — ending a step before it's done. Fix the\n  completion criterion first (cheap); only split to hide later steps if the\n  criterion is irreducibly fuzzy *and* you see the rush. When the skill's\n  job ends in a handover artifact, name it as the *only* skill-level\n  done-condition — otherwise runs end at whichever intermediate artifact\n  feels finished.\n- **Embargo** — an ordered workflow over-obeyed: the agent withholds a\n  finding made early to honor a later step's choreography, so the user\n  decides something while the agent sits on information that bears on it.\n  Any skill that sequences steps needs the escape valve stated: order\n  governs presentation, never disclosure.\n- **Lucky pass** — a validation run that succeeds only because the user or\n  world volunteered a critical input unprompted. The outcome was right but\n  the process didn't produce it; encode the eliciting probe as an explicit\n  step instead of banking on the luck recurring.\n- **Duplication** — the same meaning in two places. Keep a **single source\n  of truth**.\n- **Sediment** — stale layers that accumulate because adding feels safe.\n  Prune deliberately.\n- **War story** — a lesson written as the play-by-play of the change that\n  taught it: function names, tuned values, one bug's trajectory. Those\n  specifics date fast and bury the transferable rule. State the principle and\n  the smell to watch for; let the codebase hold the mechanics. One concrete\n  touchstone grounds it; a paragraph of them drowns it.\n- **Implementation index** — a skill that points at today's source files, line\n  numbers, current literals, or exact internal functions when its job is really\n  to teach judgment. Those locators rot and make agents chase old mechanics.\n  Keep durable principles, symptoms, acceptance criteria, and reference assets\n  in the skill; put task-specific implementation notes in the active spec. If\n  code location matters, tell the agent to find the current owner in the\n  codebase.\n- **Sprawl** — too long even when every line is live. Cure with the ladder:\n  disclose reference behind pointers, split by branch or sequence.\n- **No-op** — a line the model already obeys by default. Test each sentence\n  in isolation; when it fails, delete the whole sentence, don't trim words.\n\n## Shape\n\nUse this structure unless there is a strong reason not to:\n\n```markdown\n---\nname: short-verb-phrase\ndescription: What this does. Use when ...\n---\n\n# Skill Title\n\nOne short paragraph defining the job.\n\n## Workflow\n\n1. Do the first load-bearing thing.\n2. Make the key decision.\n3. Produce or verify the artifact.\n\n## Rules\n\n- Keep the constraints that prevent common mistakes.\n- Link only the references that should be loaded conditionally.\n```\n\n## Edit Pass\n\nWhen creating or revising a skill:\n\n- Prefer **action + plural noun** in lowercase hyphen-case: `audit-tests`,\n  `write-skills`, `compare-screenshots`. Keep the folder and frontmatter name\n  identical. Use a natural exception when pluralizing would distort the meaning;\n  this is the default for new names, not a mandate to rename existing skills.\n- Make the description specific enough to trigger without the body.\n- Remove any \"when to use\" section from the body.\n- Remove stale history, placeholders, and setup notes. Preserve attribution,\n  provenance records, and required license notices.\n- Remove file paths, line numbers, current constants, and implementation knobs\n  unless the skill is explicitly a code-navigation runbook. Prefer durable\n  principles plus a directive to inspect the current code.\n- Prefer one strong rule over several overlapping bullets.\n- Refactor restatements into a leading word where one fits.\n- Keep examples tiny and realistic.\n- Add no README, changelog, or auxiliary docs unless they are actual\n  references the skill tells the agent when to read.\n- Run the skill validator when available.\n\n## Done\n\nThe skill is done when its metadata triggers correctly, its body is short\nenough to read in one pass, and a fresh agent can follow it without asking\nwhy the skill exists.\n",{"name":5,"description":7},"main","fc9ff73613f4346442defd1318eae264af415d72","skills\u002Fwrite-skills",1,"https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit","https:\u002F\u002Ftransitivebullsh.it","My heart is open source 💕 ","2026-10-24 01:38:48",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add transitive-bullshit\u002Fskills --skill write-skills",{"tool":40,"command":41},"bunx","bunx skills add transitive-bullshit\u002Fskills --skill write-skills",{"tool":43,"command":44},"pnpm","pnpm dlx skills add transitive-bullshit\u002Fskills --skill write-skills","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwrite-skills\u002FSKILL.md",true,[48],{"path":49,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":16},"LICENSE","许可证",1068,[53,60,67,74,81,88],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":20},"01M7C8Z0JQ0VH6YYSJQGXJKVV4","xurl","Paid X API fallback through xurl after cheaper X reads fail, or for authorized account actions unsupported by bird.","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":20},"01M7C8YY74K2MJYK311R1PG31A","x-data","Read or research X\u002FTwitter posts, URLs, profiles, threads, or the user's X history using the cheapest available source.","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fx-data\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":20},"01M7C8YQFGTFVQQJCH8SPFS3YQ","web-perf","Audit, diagnose, or optimize website loading and interaction performance, Core Web Vitals, and Lighthouse performance scores.","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fweb-perf\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":20},"01M7C8YJ6QFV6QFCJESHGZZ2B8","vercel-react-best-practices","Apply React and Next.js performance guidance. Use for performance work or a specific rendering, data-fetching, or bundle concern.","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fvercel-react-best-practices\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":20},"01M7C8YBRVCMAZ99D1Y5QXHMG8","vercel-optimize","Audit deployed Vercel cost and performance using project metrics. Use for expensive or slow routes, usage analysis, and caching opportunities.","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fvercel-optimize\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":20},"01M7C8Y8Z3DGF681PACA1864E8","variant","Build distinct component alternatives and help the user choose. Use when explicitly invoked for design comparison.","https:\u002F\u002Fgithub.com\u002Ftransitive-bullshit\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fvariant\u002FSKILL.md",[],[]]