[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M69B1GZD2X0MKB22FS1RHG9S":3,"skill-same:01M69B1GZD2X0MKB22FS1RHG9S":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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M69B1GZD2X0MKB22FS1RHG9S","personafy",null,"Build a new opinionated advisor-persona skill — a reviewer \"lens\" like crusty-old-engineer, modeled on a real person or archetype. Mines the subject's authentic voice, defines its one load-bearing question, drafts it to the family template, and proves it steers a live session before publishing to a skills bundle. Use when creating or authoring a persona\u002Fadvisor skill, adding a sibling to the crusty-old-engineer family, or turning a person's real direction style into a reusable reviewer skill. Also triggers on \"personafy\" \u002F \"personify\".","microsoft\u002Famplifier-bundle-skills","microsoft","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6154722?v=4","01M698VYDXZ6JW5WH3CRV650R8","Microsoft","01M69AZYFW50CECTV9VW6B34SE",12,"MIT","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fpersonafy\u002FSKILL.md",[],[5],"2026-11-10 10:12:30",false,"---\nname: personafy\ndisable-model-invocation: true\ndescription: >-\n  Build a new opinionated advisor-persona skill — a reviewer \"lens\" like\n  crusty-old-engineer, modeled on a real person or archetype. Mines the subject's\n  authentic voice, defines its one load-bearing question, drafts it to the family\n  template, and proves it steers a live session before publishing to a skills bundle. Use\n  when creating or authoring a persona\u002Fadvisor skill, adding a sibling to the\n  crusty-old-engineer family, or turning a person's real direction style into a reusable\n  reviewer skill. Also triggers on \"personafy\" \u002F \"personify\".\nuser-invocable: true\nshortcut: personify\nmodel_role: reasoning\n---\n\n# Personafy\n\nBuild a new **advisor-persona skill** — an opinionated reviewer \"lens\" that joins an\nexisting family of sibling personas (e.g., `crusty-old-engineer`). The persona is\nmodeled on a real person or archetype, grounded in mined evidence, and enforces ONE\ndistinct recurring question.\n\n**Success artifact:** a complete, family-conformant `SKILL.md` that is (a) grounded in\nverbatim evidence, (b) *proven to steer* an LLM in a live session, (c) reduced to the\nsmallest set that still steers, and (d) loading from its target bundle in a fresh session.\n\n## Inputs\n\n- `\u003Csubject>`: (Optional) The person or archetype to model, plus pointers to evidence\n  (session corpora, transcripts, docs). If no corpus exists, derive the persona from a\n  written conceptual brief instead.\n- `\u003Cfamily>`: (Optional) The sibling persona skills to fit alongside. Defaults to the\n  advisor-persona family in `amplifier-bundle-skills` (e.g., `crusty-old-engineer`).\n\n## Steps\n\n### 1. Study the family\n\nRead the existing sibling persona skills' `SKILL.md` in full. Extract: the shared section\ntemplate, the frontmatter\u002Fmetadata shape, the tone-contract pattern (required \u002F disallowed\n\u002F style), and — critically — the single **load-bearing question** each sibling owns and how\nthey cross-reference each other.\n\n**Success criteria:** You can state each sibling's distinct question in one line and\nreproduce the shared template and metadata shape.\n\n### 2. Observe the siblings in action (optional)\n\nRun one or more existing personas live against a shared, realistic scenario — plus a\ncombined \"consensus\" run — to see how the written instructions translate into actual\nbehavior and steering.\n\n- Execution: Delegate to `self` with `context_depth=\"none\"` (one isolated run per persona).\n\n**Success criteria:** You've *seen* how each sibling's instructions become behavior, not\njust read them — enough to know what makes the steering work.\n\n### 3. Define the load-bearing lens\n\nName the ONE distinct recurring question the new persona enforces. Build a contrast table\nagainst every sibling.\n\n**Success criteria:** A one-line lens plus a contrast table showing it is genuinely\ndistinct from each sibling.\n**Rule:** if the question collapses into a sibling's, it is not a new persona — stop.\n\n### 4. Mine the voice & discipline from real evidence\n\nThe heart of the method. Gather the subject's authentic voice and decision-discipline from\nreal data (use the conceptual fallback only if no corpus exists).\n\n- **4a. Identify corpora and weight them** — e.g., the subject's own agent-directing\n  sessions (long-term backbone) and human-to-human transcripts (enrichment). Weight recent\n  material so it *informs* but doesn't *overpower* the long-term signal.\n- **4b. Deterministic extraction** — pull ONLY the subject's own words. Exclude delegated\n  sub-sessions (agent-to-agent), and flag\u002Fexclude automated eval-harness \u002F smoke-test noise.\n  Build a tracking directory + manifest + batches so the corpus never overflows your context.\n  - Execution: `bash` (jq\u002Fgrep\u002Fsed; never `cat` a huge session file — a single line can\n    exceed your context window).\n- **4c. Fan out scoped analysis agents** — one per batch, each following a shared evidence\n  brief, writing structured findings to disk and returning only a thin summary. Many small\n  agents beat one overloaded agent.\n  - Execution: Delegate (parallel), `model_role=\"research\"`.\n- **4d. Two-tier synthesis** — partial synthesizers consolidate groups of findings, then a\n  final synthesizer produces ONE profile: recurring traits ranked by cross-source\n  corroboration, verbatim catchphrases, pet peeves, decision lenses, and the top\n  \"this is them\" quotes. Quarantine AI-authored vocabulary (don't attribute it to the\n  subject). Record confidence and which batches were synthetic.\n  - Execution: Delegate, `model_role=\"reasoning\"`.\n\n**Conceptual fallback (no corpus):** derive the same profile fields from a written brief\nabout the archetype; mark everything as designed-not-mined.\n\n**Success criteria:** A consolidated profile where every load-bearing trait is backed by a\nverbatim quote (or explicitly marked as designed), ranked by corroboration, with synthetic\nnoise excluded.\n**Rule:** ground every claim in real evidence; an honest \"N\u002FA — not observed\" beats a\nfabricated trait.\n**Artifacts:** the profile file path.\n\n### 5. Draft the SKILL.md to the family template\n\nWrite the skill mirroring the siblings exactly: identity (defined by negation — \"not X, not\nY\"), When-to-Use framed as a **cross-stage lens, not a stage-gate**, the tone contract\n(required \u002F disallowed \u002F style), Core Behaviors each anchored in a **verbatim quote** from\nthe profile, an Output Structure, one worked Example, Explicit Non-Goals, a\nRelationship-to-siblings section, and a Final Note. Match the family's frontmatter format.\n\n**Success criteria:** A complete draft, structurally identical to the siblings, every\nbehavior grounded in the profile.\n**Rule:** keep the verbatim quotes — they are the highest-signal tokens and the soul of the\npersona.\n**Rule:** do NOT include `allowed-tools` in the generated SKILL.md. Persona advisor skills are\ninline (no `context: fork`), so the field is inert. If you must restrict tools for a fork-based\nvariant, use Amplifier **module IDs** (e.g. `tool-filesystem`, `tool-bash`, `tool-delegate`) —\nnever Claude Code tool names (`Read`, `Grep`, `Bash`, `Agent`).\n\n### 6. Prove it steers in a live session\n\nRun the draft in a fresh CLI session against a real-ish scenario and confirm it adopts the\nvoice and hits each behavior.\n\n```bash\namplifier run --mode single --output-format text \\\n  \"Load the skill \\\"\u003Cname>\\\" and review the following as that persona: \u003Crealistic scenario>\"\n```\n\n**Success criteria:** A transcript showing the persona *behaving* as designed (voice + each\nbehavior firing).\n**Rule:** proof is demonstrated behavior — \"the file exists\" is NOT proof.\n\n### 7. Reduce & refine\n\nApply context-reduction: cut redundancy to the smallest set that still steers (drop restated\nprocedure, compress prose), but keep the verbatim quotes and the structural skeleton.\nRe-prove (Step 6) if the cut was heavy.\n\n**Check the frontmatter `description:` length, not just the body.** The enforced cap is\n**400 characters, ERROR at 800** — `foundation:recipes\u002Fvalidate-bundle-repo.yaml` Phase 2.82,\nper `foundation:context\u002Fshared\u002Fdescription-authoring-principles.md` V5. (The Agent Skills\nspec's 1024-char ceiling is a much looser upper bound and is NOT the operative limit; the\n`tool-skills` warning past 1024 is soft and fires long after the real budget is blown.)\nEvery visible skill's full `description` is injected on every turn, so an over-long one is a\npermanently-recurring cost, not a one-time nuisance. Measure it (e.g.\n`python3 -c \"import yaml; print(len(yaml.safe_load(open('SKILL.md').read().split('---')[1])['description']))\"`)\nand if it is past 400, compress — do not relocate the trimmed content into the body,\nsince the visibility hook only ever shows `description`, never the body. The single highest-value\ncut is almost always the negation-identity clause (e.g. \"Not a long-term ownership-cost reviewer —\na reviewer of whether THIS bet, sized as proposed, is a bet the team can actually win.\" compresses\nto \"Not a cost reviewer — a bet-sizing reviewer.\" with zero loss of routing signal, since the full\nnuance already lives in the body's Identity section). Never cut the \"Use when:\" trigger list itself\n— that's the part carrying the routing weight this step must preserve.\n\n**Success criteria:** A leaner SKILL.md with the same steering power, verified, and a\n`description:` field **at or below 400 characters**. Do not calibrate against sibling\nnorms — measured 2026-09-07, 17 of this bundle's 38 skills are over that cap and 7 are\nover the 800-char ERROR line, so the siblings are the drift, not the standard. If a\nrouting fact genuinely will not fit, keep the fact, exceed the cap, and say in the PR\nwhich fact forced it: fidelity beats brevity (V7).\n\n### 8. Name it and choose its home\n\nPick a name in the family's convention. A **shareable archetype** name (a generic role) →\na public bundle; a name that **references a real person** → a private\u002Fteam bundle.\n\n- Execution: Delegate to design\u002Fvoice agents for name options (optional).\n- **Human checkpoint:** the user chooses the final name and target bundle.\n\n**Success criteria:** Name chosen and target bundle decided — public vs private matched to\nwhether the name exposes a real person.\n\n### 9. Publish to the target bundle\n\nPlace `SKILL.md` at `\u003Cbundle>\u002Fskills\u002F\u003Cname>\u002FSKILL.md`. Verify the bundle exposes skills via\nthe **canonical directory-discovery** registration — its `tool-skills` config points\n`config.skills` at the whole `skills\u002F` directory (`#subdirectory=skills`), so a new skill\ndir is auto-discovered. Do NOT rely on per-skill wrapper behaviors. Then run the git\nlifecycle.\n\n- Execution: Delegate the branch\u002Fcommit\u002FPR\u002Fmerge to `foundation:git-ops`.\n- **Human checkpoint:** confirm before opening\u002Fmerging the PR.\n\n**Success criteria:** PR merged into the target bundle.\n**Rule:** a `SKILL.md` in `skills\u002F` is NOT exposed unless the bundle points `tool-skills`\nat the `skills\u002F` directory — registration is directory-based, not per-file.\n\n### 10. Prove it loads from the bundle\n\nRun `amplifier update`, then load the skill in a FRESH session and confirm `loaded_from` is\nthe bundle cache — not a local `~\u002F.amplifier\u002Fskills` copy.\n\n**Success criteria:** `load_skill` resolves the skill from the bundle cache path in a clean\nsession.\n**Rule:** \"merged on main\" ≠ \"loads for a user\" — verify discoverability end-to-end, the\nsame gate the persona itself would demand.\n","# Personafy\n\nBuild a new **advisor-persona skill** — an opinionated reviewer \"lens\" that joins an\nexisting family of sibling personas (e.g., `crusty-old-engineer`). The persona is\nmodeled on a real person or archetype, grounded in mined evidence, and enforces ONE\ndistinct recurring question.\n\n**Success artifact:** a complete, family-conformant `SKILL.md` that is (a) grounded in\nverbatim evidence, (b) *proven to steer* an LLM in a live session, (c) reduced to the\nsmallest set that still steers, and (d) loading from its target bundle in a fresh session.\n\n## Inputs\n\n- `\u003Csubject>`: (Optional) The person or archetype to model, plus pointers to evidence\n  (session corpora, transcripts, docs). If no corpus exists, derive the persona from a\n  written conceptual brief instead.\n- `\u003Cfamily>`: (Optional) The sibling persona skills to fit alongside. Defaults to the\n  advisor-persona family in `amplifier-bundle-skills` (e.g., `crusty-old-engineer`).\n\n## Steps\n\n### 1. Study the family\n\nRead the existing sibling persona skills' `SKILL.md` in full. Extract: the shared section\ntemplate, the frontmatter\u002Fmetadata shape, the tone-contract pattern (required \u002F disallowed\n\u002F style), and — critically — the single **load-bearing question** each sibling owns and how\nthey cross-reference each other.\n\n**Success criteria:** You can state each sibling's distinct question in one line and\nreproduce the shared template and metadata shape.\n\n### 2. Observe the siblings in action (optional)\n\nRun one or more existing personas live against a shared, realistic scenario — plus a\ncombined \"consensus\" run — to see how the written instructions translate into actual\nbehavior and steering.\n\n- Execution: Delegate to `self` with `context_depth=\"none\"` (one isolated run per persona).\n\n**Success criteria:** You've *seen* how each sibling's instructions become behavior, not\njust read them — enough to know what makes the steering work.\n\n### 3. Define the load-bearing lens\n\nName the ONE distinct recurring question the new persona enforces. Build a contrast table\nagainst every sibling.\n\n**Success criteria:** A one-line lens plus a contrast table showing it is genuinely\ndistinct from each sibling.\n**Rule:** if the question collapses into a sibling's, it is not a new persona — stop.\n\n### 4. Mine the voice & discipline from real evidence\n\nThe heart of the method. Gather the subject's authentic voice and decision-discipline from\nreal data (use the conceptual fallback only if no corpus exists).\n\n- **4a. Identify corpora and weight them** — e.g., the subject's own agent-directing\n  sessions (long-term backbone) and human-to-human transcripts (enrichment). Weight recent\n  material so it *informs* but doesn't *overpower* the long-term signal.\n- **4b. Deterministic extraction** — pull ONLY the subject's own words. Exclude delegated\n  sub-sessions (agent-to-agent), and flag\u002Fexclude automated eval-harness \u002F smoke-test noise.\n  Build a tracking directory + manifest + batches so the corpus never overflows your context.\n  - Execution: `bash` (jq\u002Fgrep\u002Fsed; never `cat` a huge session file — a single line can\n    exceed your context window).\n- **4c. Fan out scoped analysis agents** — one per batch, each following a shared evidence\n  brief, writing structured findings to disk and returning only a thin summary. Many small\n  agents beat one overloaded agent.\n  - Execution: Delegate (parallel), `model_role=\"research\"`.\n- **4d. Two-tier synthesis** — partial synthesizers consolidate groups of findings, then a\n  final synthesizer produces ONE profile: recurring traits ranked by cross-source\n  corroboration, verbatim catchphrases, pet peeves, decision lenses, and the top\n  \"this is them\" quotes. Quarantine AI-authored vocabulary (don't attribute it to the\n  subject). Record confidence and which batches were synthetic.\n  - Execution: Delegate, `model_role=\"reasoning\"`.\n\n**Conceptual fallback (no corpus):** derive the same profile fields from a written brief\nabout the archetype; mark everything as designed-not-mined.\n\n**Success criteria:** A consolidated profile where every load-bearing trait is backed by a\nverbatim quote (or explicitly marked as designed), ranked by corroboration, with synthetic\nnoise excluded.\n**Rule:** ground every claim in real evidence; an honest \"N\u002FA — not observed\" beats a\nfabricated trait.\n**Artifacts:** the profile file path.\n\n### 5. Draft the SKILL.md to the family template\n\nWrite the skill mirroring the siblings exactly: identity (defined by negation — \"not X, not\nY\"), When-to-Use framed as a **cross-stage lens, not a stage-gate**, the tone contract\n(required \u002F disallowed \u002F style), Core Behaviors each anchored in a **verbatim quote** from\nthe profile, an Output Structure, one worked Example, Explicit Non-Goals, a\nRelationship-to-siblings section, and a Final Note. Match the family's frontmatter format.\n\n**Success criteria:** A complete draft, structurally identical to the siblings, every\nbehavior grounded in the profile.\n**Rule:** keep the verbatim quotes — they are the highest-signal tokens and the soul of the\npersona.\n**Rule:** do NOT include `allowed-tools` in the generated SKILL.md. Persona advisor skills are\ninline (no `context: fork`), so the field is inert. If you must restrict tools for a fork-based\nvariant, use Amplifier **module IDs** (e.g. `tool-filesystem`, `tool-bash`, `tool-delegate`) —\nnever Claude Code tool names (`Read`, `Grep`, `Bash`, `Agent`).\n\n### 6. Prove it steers in a live session\n\nRun the draft in a fresh CLI session against a real-ish scenario and confirm it adopts the\nvoice and hits each behavior.\n\n```bash\namplifier run --mode single --output-format text \\\n  \"Load the skill \\\"\u003Cname>\\\" and review the following as that persona: \u003Crealistic scenario>\"\n```\n\n**Success criteria:** A transcript showing the persona *behaving* as designed (voice + each\nbehavior firing).\n**Rule:** proof is demonstrated behavior — \"the file exists\" is NOT proof.\n\n### 7. Reduce & refine\n\nApply context-reduction: cut redundancy to the smallest set that still steers (drop restated\nprocedure, compress prose), but keep the verbatim quotes and the structural skeleton.\nRe-prove (Step 6) if the cut was heavy.\n\n**Check the frontmatter `description:` length, not just the body.** The enforced cap is\n**400 characters, ERROR at 800** — `foundation:recipes\u002Fvalidate-bundle-repo.yaml` Phase 2.82,\nper `foundation:context\u002Fshared\u002Fdescription-authoring-principles.md` V5. (The Agent Skills\nspec's 1024-char ceiling is a much looser upper bound and is NOT the operative limit; the\n`tool-skills` warning past 1024 is soft and fires long after the real budget is blown.)\nEvery visible skill's full `description` is injected on every turn, so an over-long one is a\npermanently-recurring cost, not a one-time nuisance. Measure it (e.g.\n`python3 -c \"import yaml; print(len(yaml.safe_load(open('SKILL.md').read().split('---')[1])['description']))\"`)\nand if it is past 400, compress — do not relocate the trimmed content into the body,\nsince the visibility hook only ever shows `description`, never the body. The single highest-value\ncut is almost always the negation-identity clause (e.g. \"Not a long-term ownership-cost reviewer —\na reviewer of whether THIS bet, sized as proposed, is a bet the team can actually win.\" compresses\nto \"Not a cost reviewer — a bet-sizing reviewer.\" with zero loss of routing signal, since the full\nnuance already lives in the body's Identity section). Never cut the \"Use when:\" trigger list itself\n— that's the part carrying the routing weight this step must preserve.\n\n**Success criteria:** A leaner SKILL.md with the same steering power, verified, and a\n`description:` field **at or below 400 characters**. Do not calibrate against sibling\nnorms — measured 2026-09-07, 17 of this bundle's 38 skills are over that cap and 7 are\nover the 800-char ERROR line, so the siblings are the drift, not the standard. If a\nrouting fact genuinely will not fit, keep the fact, exceed the cap, and say in the PR\nwhich fact forced it: fidelity beats brevity (V7).\n\n### 8. Name it and choose its home\n\nPick a name in the family's convention. A **shareable archetype** name (a generic role) →\na public bundle; a name that **references a real person** → a private\u002Fteam bundle.\n\n- Execution: Delegate to design\u002Fvoice agents for name options (optional).\n- **Human checkpoint:** the user chooses the final name and target bundle.\n\n**Success criteria:** Name chosen and target bundle decided — public vs private matched to\nwhether the name exposes a real person.\n\n### 9. Publish to the target bundle\n\nPlace `SKILL.md` at `\u003Cbundle>\u002Fskills\u002F\u003Cname>\u002FSKILL.md`. Verify the bundle exposes skills via\nthe **canonical directory-discovery** registration — its `tool-skills` config points\n`config.skills` at the whole `skills\u002F` directory (`#subdirectory=skills`), so a new skill\ndir is auto-discovered. Do NOT rely on per-skill wrapper behaviors. Then run the git\nlifecycle.\n\n- Execution: Delegate the branch\u002Fcommit\u002FPR\u002Fmerge to `foundation:git-ops`.\n- **Human checkpoint:** confirm before opening\u002Fmerging the PR.\n\n**Success criteria:** PR merged into the target bundle.\n**Rule:** a `SKILL.md` in `skills\u002F` is NOT exposed unless the bundle points `tool-skills`\nat the `skills\u002F` directory — registration is directory-based, not per-file.\n\n### 10. Prove it loads from the bundle\n\nRun `amplifier update`, then load the skill in a FRESH session and confirm `loaded_from` is\nthe bundle cache — not a local `~\u002F.amplifier\u002Fskills` copy.\n\n**Success criteria:** `load_skill` resolves the skill from the bundle cache path in a clean\nsession.\n**Rule:** \"merged on main\" ≠ \"loads for a user\" — verify discoverability end-to-end, the\nsame gate the persona itself would demand.\n",{"name":5,"shortcut":24,"model_role":25,"description":7,"user-invocable":26,"disable-model-invocation":26},"personify","reasoning",true,"main","2f58408a879535df36da5c92d2b073f49271ade4","skills\u002Fpersonafy",11,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills","https:\u002F\u002Fgithub.com\u002Fmicrosoft","https:\u002F\u002Fopensource.microsoft.com","Open source projects and samples from Microsoft","2026-10-03 02:23:42",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":26},[38,41,44],{"tool":39,"command":40},"npx","npx skills add microsoft\u002Famplifier-bundle-skills --skill personafy",{"tool":42,"command":43},"bunx","bunx skills add microsoft\u002Famplifier-bundle-skills --skill personafy",{"tool":45,"command":46},"pnpm","pnpm dlx skills add microsoft\u002Famplifier-bundle-skills --skill personafy","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fpersonafy\u002FSKILL.md",[49,55,59,63,67,71,75,79,83,87],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fbeats.md","REFERENCE","参考文档",921,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fbeats.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fcloud.md",11119,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fcloud.md",{"path":60,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fcloudrun.md",3583,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fcloudrun.md",{"path":64,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":65,"sourceHtmlUrl":66},"references\u002Fcompare-and-batch.md",4378,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fcompare-and-batch.md",{"path":68,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":69,"sourceHtmlUrl":70},"references\u002Fdoctor-browser.md",4357,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fdoctor-browser.md",{"path":72,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":73,"sourceHtmlUrl":74},"references\u002Finit-and-scaffold.md",5798,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Finit-and-scaffold.md",{"path":76,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":77,"sourceHtmlUrl":78},"references\u002Flambda.md",16703,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Flambda.md",{"path":80,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Flint-validate-inspect.md",13424,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Flint-validate-inspect.md",{"path":84,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Fpreview-render.md",31668,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fpreview-render.md",{"path":88,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Fupgrade-info-misc.md",11389,"https:\u002F\u002Fgithub.com\u002Fmdnaimul22\u002Fhuman-skills\u002Fblob\u002Fmain\u002Fskills\u002Fstorage\u002Fheygen-com_hyperframes\u002Fhyperframes-cli\u002Freferences\u002Fupgrade-info-misc.md",[92,102,113,126,137,151],{"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":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M69B282M59QZNMM26JXA1QCV","verification-discipline","Use when verifying that completed work actually works. Auto-surface during \u002Fverify mode, post-implementation review, or before claiming a task is done. Teaches testing outcomes vs implementation, the unit\u002Fintegration\u002Fsmoke gradient, and what \"done\" actually means.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fverification-discipline\u002FSKILL.md",[],[99,100,101],"verification discipline","review","testing",{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":19},"01M69B263M6BFCVAV3SEESV7T0","user-advocate","User-need reviewer that speaks for the person who isn't in the room — the one who will actually live with what gets built. Hunts the gap between \"we can build this\" and \"they actually want this,\" in the patient, impatient voice of the absent user. Use at any checkpoint — brainstorm, design, plan, implement, debug, review — when a feature is being built because it's buildable rather than wanted, the recovery path is missing, or nobody can name the person this serves: \"do they actually want this, and can they live with it?\"","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fuser-advocate\u002FSKILL.md",[],[109,110,111,100,112],"user advocate","debug","implement","design",{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":124,"publishedAt":19},"01M69B23EQQM30TMZG7FPJW4F7","tester-breaker","Adversarial breaker that reviews code by trying to make it fail, not by confirming it works. Hunts the unhappy paths — malformed input, the empty string, the reversed range, the race condition — as a gleeful adversary who assumes everything is broken until a concrete attempt to break it comes up empty. Use at any checkpoint — brainstorm, design, plan, implement, debug, review — when the happy path is celebrated while the edges sit unexamined, \"looks fine\" stands in for \"I tried to break it and couldn't,\" or nobody has named the input that makes this fall over: \"how does this fail, and where are the edges?\"","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Ftester-breaker\u002FSKILL.md",[119],{"categoryNo":120,"categoryName":121,"taxonomyType":122,"taxonomyTypeDesc":123},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[125,100,110,111,112],"tester breaker",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":19},"01M69B21AHBXP6EK60XYDP24PY","skills-assist","Authoritative consultant for all skills-related questions. Use when creating or modifying skills, understanding the Agent Skills spec, troubleshooting skill loading or invocation issues, leveraging enhanced format features (context fork, model_role, user-invocable), writing cross-harness portable skills, ensuring Claude Code Skills 2.0 compatibility, or deciding between skills vs agents.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fskills-assist\u002FSKILL.md",[],[133,134,135,136],"skills assist","claude code skills","spec","writing",{"skillNo":138,"name":139,"nameZh":6,"description":140,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":141,"categories":142,"keywords":148,"publishedAt":19},"01M69B1ZBHGE1PAX0E7ZDS8HQZ","skillify","Capture a repeatable process from the current session into a reusable Amplifier SKILL.md skill file. Use when the user wants to create a skill, save a workflow as a skill, turn a process into a reusable skill, or mentions \"skillify\", \"create skill\", \"make a skill\", \"save as skill\", \"capture workflow\", \"turn this into a skill\", \"new skill\", or wants to automate a repeatable process they just performed.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fskillify\u002FSKILL.md",[143],{"categoryNo":144,"categoryName":145,"taxonomyType":146,"taxonomyTypeDesc":147},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[139,149,150],"create skill","workflow",{"skillNo":152,"name":153,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":158,"publishedAt":19},"01M69B1XCPQFAXS9RTS87CZ8EK","session-debug","Diagnose issues in the current Amplifier session — misconfigured tools, failing operations, unexpected behavior. Use when something isn't working right.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Famplifier-bundle-skills\u002Fblob\u002Fmain\u002Fskills\u002Fsession-debug\u002FSKILL.md",[157],{"categoryNo":120,"categoryName":121,"taxonomyType":122,"taxonomyTypeDesc":123},[159,110,160],"session debug","diagnose"]