[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6D3PT65G4M7MMVGA41HNTP7":3,"skill-same:01M6D3PT65G4M7MMVGA41HNTP7":65},{"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":26,"linkOnly":27,"markdownContent":28,"markdownBody":29,"frontMatter":30,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":40,"ownerHtmlUrl":41,"ownerBlogUrl":42,"ownerDescription":43,"updatedAt":44,"install":45,"resources":58},"01M6D3PT65G4M7MMVGA41HNTP7","autoresearch-prep",null,"Scaffold a program.md for $autoresearch by auto-detecting codebase signals and interviewing for what can't be inferred","GeorgeQLe\u002Fagentic-skills","GeorgeQLe","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F21134126?v=4","01M6D2JP418S0VGJGF8M93EC3H","George Le","01M6D2JPBAKYD82TA5FS6HKQ3D",1,"MIT","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Fpacks\u002Fbase\u002Fcodex\u002Fautoresearch-prep\u002FSKILL.md",[],[19,20,21,22,23,24,25],"autoresearch","autoresearch prep","scaffold","planning","v0","auto","signals","2026-11-10 10:12:30",false,"---\nname: autoresearch-prep\ndescription: Scaffold a program.md for $autoresearch by auto-detecting codebase signals and interviewing for what can't be inferred\ntype: planning\nversion: v0.0\ninvocation: chained\nargument-hint: \"[output-path]\"\ncontext_intake: scoped\n---\n\n# Autoresearch Prep\n\nInvoke as `$autoresearch-prep`.\n\nScaffolds a `program.md` research program for `$autoresearch`. Auto-detects benchmark commands, test commands, sandbox boundaries, and exclude patterns from the codebase, then asks the user only what it can't infer: metric command and direction, research directions, and budget.\n\n## When to Use\n\nWhen the user wants to start an `$autoresearch` loop but doesn't have a `program.md` yet.\n\n## Preconditions\n\n- A git repository with code to optimize\n- The user has a measurable property in mind (benchmark, bundle size, test coverage, etc.)\n\n## Process\n\n### 0. Resolve Output Path\n\n1. If `$ARGUMENTS` is provided and non-empty, use it as the output path.\n2. Otherwise default to `.\u002Fprogram.md`.\n3. If the file already exists, warn the user and ask whether to overwrite or pick a new path.\n\n### 1. Auto-Detect Codebase Signals\n\nScan the repository for:\n\n**Benchmark commands:**\n- `package.json` scripts containing `bench` or `benchmark`\n- `bench\u002F`, `benchmarks\u002F`, or `perf\u002F` directories\n- `Makefile` targets: `bench`, `benchmark`, `perf`\n- Language-specific: `cargo bench`, `go test -bench`, `pytest-benchmark`, `hyperfine` usage\n\n**Test commands:**\n- `package.json` `test` script\n- `Makefile` `test` target\n- Language-specific: `pytest`, `cargo test`, `go test`, `jest`, `vitest`, `mocha`\n\n**Sandbox globs** (based on what directories exist):\n- `src\u002F**`, `lib\u002F**`, `app\u002F**`, `pkg\u002F**`, `internal\u002F**`, `crates\u002F**`\n- Include only directories that actually exist in the repo\n\n**Exclude patterns:**\n- Test files: `**\u002F*.test.*`, `**\u002F*.spec.*`, `**\u002Ftest\u002F**`, `**\u002Ftests\u002F**`, `**\u002F__tests__\u002F**`\n- Config: `*.config.*`, `.eslintrc*`, `tsconfig*`, `.prettierrc*`\n- Benchmark harness: `bench\u002F**`, `benchmarks\u002F**`, `perf\u002F**`\n- Build output: `dist\u002F**`, `build\u002F**`, `target\u002F**`, `out\u002F**`, `.next\u002F**`\n- Lock files: `package-lock.json`, `yarn.lock`, `pnpm-lock.yaml`, `Cargo.lock`, `go.sum`\n- VCS\u002Ftooling: `.git\u002F**`, `node_modules\u002F**`, `.autoresearch\u002F**`\n\n### 2. Light Interview\n\nPresent the detected signals as a summary, then ask:\n\n1. **Metric command + direction** (always required): \"What shell command prints your target metric as a single number? Is higher or lower better?\"\n2. **Research directions** (always required): \"What optimization directions should the agent explore? List 2-5 ideas.\"\n3. **Budget** (offer default): \"How many iterations? (default: 20)\"\n\nIf benchmark commands were detected, suggest them as starting points for the metric command.\nIf test commands were detected, note they'll be included as the test command with `required: true`.\n\n### 3. Findings Validation\n\nShow the complete proposed `program.md` content to the user. Let them confirm or adjust any section before writing.\n\n### 4. Validate Metric Command\n\n1. Run the metric command once from the project root.\n2. Parse the last number from stdout.\n3. If the command fails or produces no parseable number, stop and ask the user to fix it.\n4. Record the parsed value as the baseline for the Context section.\n\n### 5. Write program.md\n\nWrite the file in the exact format expected by `$autoresearch`:\n\n```markdown\n# Research Program\n\n## Metric\ncommand: \u003Cuser-provided command>\ndirection: \u003Chigher-is-better | lower-is-better>\n\n## Budget\nmax_iterations: \u003Cuser-provided or 20>\n\n## Sandbox\nfiles:\n  - \u003Cdetected or user-adjusted globs>\nexclude:\n  - \u003Cdetected or user-adjusted patterns>\n\n## Test Command\ncommand: \u003Cdetected test command>\nrequired: true\n\n## Research Directions\n1. \u003Cdirection one>\n2. \u003Cdirection two>\n...\n\n## Context\nBaseline metric value: \u003Cmeasured value>\n\u003Cany additional context from interview>\n```\n\nOmit the `Test Command` section if no test command was detected or provided.\n\n### 6. Route\n\n**Recommended next command:** `$autoresearch \u003Coutput-path>`\n\n## Constraints\n\n- Do not run the autoresearch loop — only scaffold `program.md`.\n- Do not modify any source files in the repository.\n- Do not create the `.autoresearch\u002F` directory.\n- Do not create alignment pages.\n- Do not follow the shared shipping contract — this skill writes one local file, not tracked repo changes.\n","# Autoresearch Prep\n\nInvoke as `$autoresearch-prep`.\n\nScaffolds a `program.md` research program for `$autoresearch`. Auto-detects benchmark commands, test commands, sandbox boundaries, and exclude patterns from the codebase, then asks the user only what it can't infer: metric command and direction, research directions, and budget.\n\n## When to Use\n\nWhen the user wants to start an `$autoresearch` loop but doesn't have a `program.md` yet.\n\n## Preconditions\n\n- A git repository with code to optimize\n- The user has a measurable property in mind (benchmark, bundle size, test coverage, etc.)\n\n## Process\n\n### 0. Resolve Output Path\n\n1. If `$ARGUMENTS` is provided and non-empty, use it as the output path.\n2. Otherwise default to `.\u002Fprogram.md`.\n3. If the file already exists, warn the user and ask whether to overwrite or pick a new path.\n\n### 1. Auto-Detect Codebase Signals\n\nScan the repository for:\n\n**Benchmark commands:**\n- `package.json` scripts containing `bench` or `benchmark`\n- `bench\u002F`, `benchmarks\u002F`, or `perf\u002F` directories\n- `Makefile` targets: `bench`, `benchmark`, `perf`\n- Language-specific: `cargo bench`, `go test -bench`, `pytest-benchmark`, `hyperfine` usage\n\n**Test commands:**\n- `package.json` `test` script\n- `Makefile` `test` target\n- Language-specific: `pytest`, `cargo test`, `go test`, `jest`, `vitest`, `mocha`\n\n**Sandbox globs** (based on what directories exist):\n- `src\u002F**`, `lib\u002F**`, `app\u002F**`, `pkg\u002F**`, `internal\u002F**`, `crates\u002F**`\n- Include only directories that actually exist in the repo\n\n**Exclude patterns:**\n- Test files: `**\u002F*.test.*`, `**\u002F*.spec.*`, `**\u002Ftest\u002F**`, `**\u002Ftests\u002F**`, `**\u002F__tests__\u002F**`\n- Config: `*.config.*`, `.eslintrc*`, `tsconfig*`, `.prettierrc*`\n- Benchmark harness: `bench\u002F**`, `benchmarks\u002F**`, `perf\u002F**`\n- Build output: `dist\u002F**`, `build\u002F**`, `target\u002F**`, `out\u002F**`, `.next\u002F**`\n- Lock files: `package-lock.json`, `yarn.lock`, `pnpm-lock.yaml`, `Cargo.lock`, `go.sum`\n- VCS\u002Ftooling: `.git\u002F**`, `node_modules\u002F**`, `.autoresearch\u002F**`\n\n### 2. Light Interview\n\nPresent the detected signals as a summary, then ask:\n\n1. **Metric command + direction** (always required): \"What shell command prints your target metric as a single number? Is higher or lower better?\"\n2. **Research directions** (always required): \"What optimization directions should the agent explore? List 2-5 ideas.\"\n3. **Budget** (offer default): \"How many iterations? (default: 20)\"\n\nIf benchmark commands were detected, suggest them as starting points for the metric command.\nIf test commands were detected, note they'll be included as the test command with `required: true`.\n\n### 3. Findings Validation\n\nShow the complete proposed `program.md` content to the user. Let them confirm or adjust any section before writing.\n\n### 4. Validate Metric Command\n\n1. Run the metric command once from the project root.\n2. Parse the last number from stdout.\n3. If the command fails or produces no parseable number, stop and ask the user to fix it.\n4. Record the parsed value as the baseline for the Context section.\n\n### 5. Write program.md\n\nWrite the file in the exact format expected by `$autoresearch`:\n\n```markdown\n# Research Program\n\n## Metric\ncommand: \u003Cuser-provided command>\ndirection: \u003Chigher-is-better | lower-is-better>\n\n## Budget\nmax_iterations: \u003Cuser-provided or 20>\n\n## Sandbox\nfiles:\n  - \u003Cdetected or user-adjusted globs>\nexclude:\n  - \u003Cdetected or user-adjusted patterns>\n\n## Test Command\ncommand: \u003Cdetected test command>\nrequired: true\n\n## Research Directions\n1. \u003Cdirection one>\n2. \u003Cdirection two>\n...\n\n## Context\nBaseline metric value: \u003Cmeasured value>\n\u003Cany additional context from interview>\n```\n\nOmit the `Test Command` section if no test command was detected or provided.\n\n### 6. Route\n\n**Recommended next command:** `$autoresearch \u003Coutput-path>`\n\n## Constraints\n\n- Do not run the autoresearch loop — only scaffold `program.md`.\n- Do not modify any source files in the repository.\n- Do not create the `.autoresearch\u002F` directory.\n- Do not create alignment pages.\n- Do not follow the shared shipping contract — this skill writes one local file, not tracked repo changes.\n",{"name":5,"type":22,"version":31,"invocation":32,"description":7,"argument-hint":33,"context_intake":34},"v0.0","chained","[output-path]","scoped","master","96e2912af886dd1606b1b97df707715aef062b02","packs\u002Fbase\u002Fcodex\u002Fautoresearch-prep",0,"https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills","https:\u002F\u002Fagentic-skills-blush.vercel.app","https:\u002F\u002Fgithub.com\u002FGeorgeQLe","georgeqle.com","George Le is a futurist and serial tech entrepreneur who wants to work toward a world where humanity is free from robotic tasks.","2026-11-10 04:07:24",{"source":8,"skillName":5,"commands":46,"prompt":56,"zipAvailable":57},[47,50,53],{"tool":48,"command":49},"npx","npx skills add GeorgeQLe\u002Fagentic-skills --skill autoresearch-prep",{"tool":51,"command":52},"bunx","bunx skills add GeorgeQLe\u002Fagentic-skills --skill autoresearch-prep",{"tool":54,"command":55},"pnpm","pnpm dlx skills add GeorgeQLe\u002Fagentic-skills --skill autoresearch-prep","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Fpacks\u002Fbase\u002Fcodex\u002Fautoresearch-prep\u002FSKILL.md",true,[59],{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"CHANGELOG.md","REFERENCE","参考文档",146,"https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Fpacks\u002Fbase\u002Fcodex\u002Fautoresearch-prep\u002FCHANGELOG.md",[66,78,88,100,115,126],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":26},"01M6G0NJES15RJKZE6YZ0TSYGF","source-checkout-allowed","Valid source-checkout-only fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fsource-checkout-allowed\u002FSKILL.md",[],[73,74,75,76,23,77],"source checkout allowed","checkout","fixture","only","valid",{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":26},"01M6G0NFZ3R1MNB1E3DVTM0WB7","deck-install-valid","Valid deck install fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fdeck-install\u002FSKILL.md",[],[85,86,87,77,75,23],"deck install valid","deck","install",{"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":26},"01M6G0ND3QZQC4MDZ9NVB97PT2","codex-pack-guard-valid","Valid Codex Pack Availability Guard fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fcodex-pack-guard\u002FSKILL.md",[],[95,96,97,77,98,99,75,23],"codex pack guard valid","pack","guard","codex","availability",{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":111,"publishedAt":26},"01M6G0NA9YB3NKGZKD3S96JMNB","codex-missing-skill-valid","Valid Codex missing skill fallback fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fcodex-missing-skill-fallback\u002FSKILL.md",[106],{"categoryNo":107,"categoryName":108,"taxonomyType":109,"taxonomyTypeDesc":110},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[112,113,77,98,23,114,75],"codex missing skill valid","missing","fallback",{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":122,"publishedAt":26},"01M6G0N7PC0EKFD40ZWQAXSAQ9","codex-base-skill-fallback-valid","Valid Codex base skill fallback fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fcodex-base-skill-fallback\u002FSKILL.md",[121],{"categoryNo":107,"categoryName":108,"taxonomyType":109,"taxonomyTypeDesc":110},[77,98,114,123,124,125,23,75],"base skill","codex base skill fallback valid","base",{"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":26},"01M6G0N4VRR2YKZQ6N3Y6S2BQA","claude-pack-guard-valid","Valid Claude Pack Availability Guard fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fclaude-pack-guard\u002FSKILL.md",[],[77,97,96,133,99,23,75],"claude pack guard valid"]