[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CZ5NP04132733PFM8P3BJA":3,"skill-same:01M6CZ5NP04132733PFM8P3BJA":62},{"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":23,"publishedAt":29,"linkOnly":30,"markdownContent":31,"markdownBody":32,"frontMatter":33,"refName":39,"commitSha":40,"skillDirectory":41,"forksCount":42,"repositoryHtmlUrl":43,"repositoryHomepageUrl":44,"ownerHtmlUrl":45,"ownerBlogUrl":44,"ownerDescription":46,"updatedAt":47,"install":48,"resources":61},"01M6CZ5NP04132733PFM8P3BJA","autoresearch",null,"Evolve a skill by generating variations, evaluating them, and updating the best version","aeonfun\u002Faeon","aeonfun","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F287715280?v=4","01M6CZ23C6MXGMF4N0V1DZ1247","aeon","01M6CZ23GMZP1BR424J8Y8K3DG",793,"MIT","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fskills\u002Fautoresearch\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[5,24,25,26,27,28],"dev","Meta","evolve","evolution","version","2026-11-10 10:12:30",false,"---\nname: autoresearch\ndescription: Evolve a skill by generating variations, evaluating them, and updating the best version\nmetadata:\n  title: Autoresearch\n  category: evolution\n  var: \"\"\n  tags:\n    - meta\n    - dev\n---\n> **${var}** — Name of the skill to evolve (e.g. `token-movers`). Required.\n\nIf `${var}` is empty, abort with: \"autoresearch requires var= set to a skill name\" and exit.\n\nRead memory\u002FMEMORY.md for context.\n\n## Goal\n\nImprove an existing skill by researching better approaches, generating 4 distinct variations, scoring them against a rubric, and committing the winning version as a PR.\n\n## Steps\n\n### 1. Load the target skill\n\nRead `skills\u002F${var}\u002FSKILL.md`. If the file doesn't exist, abort and notify: \"Skill '${var}' not found.\"\n\nParse the skill's:\n- **Purpose**: what it does\n- **Data sources**: APIs, URLs, commands it calls\n- **Output format**: what it produces (article, notification, file)\n- **Dependencies**: env vars, tools, other files it reads\n\nSave the original content — you'll need it for the PR diff later.\n\n### 2. Research improvements\n\nSearch the web for better approaches to what this skill does:\n- Alternative or complementary APIs\u002Fdata sources\n- Best practices for the skill's domain (e.g., crypto analysis, RSS aggregation, security scanning)\n- Common pitfalls or failure modes for the techniques the skill uses\n- Output formats that are more actionable or readable\n\nAlso review:\n- Recent memory\u002Flogs\u002F entries where this skill ran — did it produce useful output? Were there failures?\n- `memory\u002Fcron-state.json` — has this skill been failing?\n\n### 3. Generate 4 variations\n\nCreate 4 distinct improved versions of the SKILL.md, each with a different thesis:\n\n**Variation A — Better inputs**: Improve data sources. Add alternative\u002Fcomplementary APIs, better search queries, more reliable endpoints. Fix any broken or deprecated sources found in step 2.\n\n**Variation B — Sharper output**: Improve the output format and content quality. Make notifications more actionable, articles more substantive, analysis more insightful. Reduce noise, improve signal.\n\n**Variation C — More robust**: Improve reliability and edge-case handling. Add fallback logic for when APIs fail, better deduplication, graceful handling of empty data, clearer error messages.\n\n**Variation D — Rethink**: Take a fundamentally different approach to achieving the same goal. Different methodology, different angle, or a creative combination of techniques the original didn't consider.\n\nEach variation must:\n- Preserve the original frontmatter format (name, description, var, tags)\n- Follow Aeon skill conventions (read memory, log to memory\u002Flogs\u002F${today}.md, notify via `.\u002Fnotify`)\n- Be a complete, ready-to-run SKILL.md — no placeholders\n- Include a one-line comment at the top of the body: `\u003C!-- autoresearch: variation X — thesis description -->`\n\n### 4. Evaluate and score\n\nScore each variation on a 1-5 scale across these criteria:\n\n| Criterion | What to evaluate |\n|-----------|-----------------|\n| **Clarity** | Will Claude execute this correctly? Are instructions unambiguous? |\n| **Data quality** | Are sources reliable, diverse, and likely to return useful data? |\n| **Output value** | Is the output actionable and worth reading? Low noise? |\n| **Robustness** | Does it handle failures, empty data, and edge cases? |\n| **Conventions** | Does it follow Aeon patterns? (memory, logging, notify, var usage) |\n| **Improvement** | How much better is this than the original? |\n\nWrite out your scoring with brief justification for each score. Calculate a weighted total:\n- Improvement: 3x weight (the whole point)\n- Output value: 2x weight\n- Clarity, Data quality, Robustness: 1.5x weight each\n- Conventions: 1x weight\n\n### 5. Select and apply the winner\n\nPick the highest-scoring variation. If scores are very close (within 2 points total), prefer the variation that makes the biggest single improvement rather than small incremental changes.\n\nWrite the winning variation to `skills\u002F${var}\u002FSKILL.md`, replacing the original.\n\n### 6. Create a PR\n\nCreate a branch named `autoresearch\u002F${var}` and commit the change:\n```bash\ngit checkout -b autoresearch\u002F${var}\ngit add skills\u002F${var}\u002FSKILL.md\ngit commit -m \"improve(${var}): autoresearch evolution\n\nVariation chosen: [A\u002FB\u002FC\u002FD] — [thesis]\nKey changes: [1-2 sentence summary]\"\ngit push -u origin autoresearch\u002F${var}\n```\n\nOpen a PR with:\n- **Title**: `improve(${var}): autoresearch evolution`\n- **Body**: Include the full scoring table, the winning variation's thesis, and a diff summary of what changed. Include all 4 variation summaries so the reviewer can see what was considered.\n\n```bash\ngh pr create --title \"improve(${var}): autoresearch evolution\" --body \"...\"\n```\n\n### 7. Notify and log\n\nSend via `.\u002Fnotify`:\n```\n*Autoresearch - ${var}*\nWinner: Variation [X] - [thesis]\nScore: [total]\u002F50\nKey changes: [summary]\nPR: [url]\n```\n\nLog to `memory\u002Flogs\u002F${today}.md`:\n```\n### autoresearch\n- Target: ${var}\n- Winner: Variation [X] ([score]\u002F50)\n- Thesis: [description]\n- PR: [url]\n- Runners-up: [brief scores]\n```\n\n## Network note\n\nThere is no network sandbox — `curl` works, with **WebFetch** as the fallback for a flaky public GET. For an auth'd API, call `.\u002Fsecretcurl` with a `{ENV_NAME}` placeholder (the key is injected via `requires:`), never a bare `$SECRET`.\n\n## Constraints\n\n- Never downgrade a working skill. If all variations score lower than or equal to the original on \"Improvement\", skip the update and notify: \"No improvement found for ${var} — all variations scored at baseline.\"\n- Preserve the skill's core purpose — evolution, not replacement.\n- Do not change the skill's tags or var semantics without strong justification.\n- Do not add env vars that aren't already available in the workflow (check aeon.yml secrets).\n","> **${var}** — Name of the skill to evolve (e.g. `token-movers`). Required.\n\nIf `${var}` is empty, abort with: \"autoresearch requires var= set to a skill name\" and exit.\n\nRead memory\u002FMEMORY.md for context.\n\n## Goal\n\nImprove an existing skill by researching better approaches, generating 4 distinct variations, scoring them against a rubric, and committing the winning version as a PR.\n\n## Steps\n\n### 1. Load the target skill\n\nRead `skills\u002F${var}\u002FSKILL.md`. If the file doesn't exist, abort and notify: \"Skill '${var}' not found.\"\n\nParse the skill's:\n- **Purpose**: what it does\n- **Data sources**: APIs, URLs, commands it calls\n- **Output format**: what it produces (article, notification, file)\n- **Dependencies**: env vars, tools, other files it reads\n\nSave the original content — you'll need it for the PR diff later.\n\n### 2. Research improvements\n\nSearch the web for better approaches to what this skill does:\n- Alternative or complementary APIs\u002Fdata sources\n- Best practices for the skill's domain (e.g., crypto analysis, RSS aggregation, security scanning)\n- Common pitfalls or failure modes for the techniques the skill uses\n- Output formats that are more actionable or readable\n\nAlso review:\n- Recent memory\u002Flogs\u002F entries where this skill ran — did it produce useful output? Were there failures?\n- `memory\u002Fcron-state.json` — has this skill been failing?\n\n### 3. Generate 4 variations\n\nCreate 4 distinct improved versions of the SKILL.md, each with a different thesis:\n\n**Variation A — Better inputs**: Improve data sources. Add alternative\u002Fcomplementary APIs, better search queries, more reliable endpoints. Fix any broken or deprecated sources found in step 2.\n\n**Variation B — Sharper output**: Improve the output format and content quality. Make notifications more actionable, articles more substantive, analysis more insightful. Reduce noise, improve signal.\n\n**Variation C — More robust**: Improve reliability and edge-case handling. Add fallback logic for when APIs fail, better deduplication, graceful handling of empty data, clearer error messages.\n\n**Variation D — Rethink**: Take a fundamentally different approach to achieving the same goal. Different methodology, different angle, or a creative combination of techniques the original didn't consider.\n\nEach variation must:\n- Preserve the original frontmatter format (name, description, var, tags)\n- Follow Aeon skill conventions (read memory, log to memory\u002Flogs\u002F${today}.md, notify via `.\u002Fnotify`)\n- Be a complete, ready-to-run SKILL.md — no placeholders\n- Include a one-line comment at the top of the body: `\u003C!-- autoresearch: variation X — thesis description -->`\n\n### 4. Evaluate and score\n\nScore each variation on a 1-5 scale across these criteria:\n\n| Criterion | What to evaluate |\n|-----------|-----------------|\n| **Clarity** | Will Claude execute this correctly? Are instructions unambiguous? |\n| **Data quality** | Are sources reliable, diverse, and likely to return useful data? |\n| **Output value** | Is the output actionable and worth reading? Low noise? |\n| **Robustness** | Does it handle failures, empty data, and edge cases? |\n| **Conventions** | Does it follow Aeon patterns? (memory, logging, notify, var usage) |\n| **Improvement** | How much better is this than the original? |\n\nWrite out your scoring with brief justification for each score. Calculate a weighted total:\n- Improvement: 3x weight (the whole point)\n- Output value: 2x weight\n- Clarity, Data quality, Robustness: 1.5x weight each\n- Conventions: 1x weight\n\n### 5. Select and apply the winner\n\nPick the highest-scoring variation. If scores are very close (within 2 points total), prefer the variation that makes the biggest single improvement rather than small incremental changes.\n\nWrite the winning variation to `skills\u002F${var}\u002FSKILL.md`, replacing the original.\n\n### 6. Create a PR\n\nCreate a branch named `autoresearch\u002F${var}` and commit the change:\n```bash\ngit checkout -b autoresearch\u002F${var}\ngit add skills\u002F${var}\u002FSKILL.md\ngit commit -m \"improve(${var}): autoresearch evolution\n\nVariation chosen: [A\u002FB\u002FC\u002FD] — [thesis]\nKey changes: [1-2 sentence summary]\"\ngit push -u origin autoresearch\u002F${var}\n```\n\nOpen a PR with:\n- **Title**: `improve(${var}): autoresearch evolution`\n- **Body**: Include the full scoring table, the winning variation's thesis, and a diff summary of what changed. Include all 4 variation summaries so the reviewer can see what was considered.\n\n```bash\ngh pr create --title \"improve(${var}): autoresearch evolution\" --body \"...\"\n```\n\n### 7. Notify and log\n\nSend via `.\u002Fnotify`:\n```\n*Autoresearch - ${var}*\nWinner: Variation [X] - [thesis]\nScore: [total]\u002F50\nKey changes: [summary]\nPR: [url]\n```\n\nLog to `memory\u002Flogs\u002F${today}.md`:\n```\n### autoresearch\n- Target: ${var}\n- Winner: Variation [X] ([score]\u002F50)\n- Thesis: [description]\n- PR: [url]\n- Runners-up: [brief scores]\n```\n\n## Network note\n\nThere is no network sandbox — `curl` works, with **WebFetch** as the fallback for a flaky public GET. For an auth'd API, call `.\u002Fsecretcurl` with a `{ENV_NAME}` placeholder (the key is injected via `requires:`), never a bare `$SECRET`.\n\n## Constraints\n\n- Never downgrade a working skill. If all variations score lower than or equal to the original on \"Improvement\", skip the update and notify: \"No improvement found for ${var} — all variations scored at baseline.\"\n- Preserve the skill's core purpose — evolution, not replacement.\n- Do not change the skill's tags or var semantics without strong justification.\n- Do not add env vars that aren't already available in the workflow (check aeon.yml secrets).\n",{"name":5,"metadata":34,"description":7},{"var":35,"tags":36,"title":38,"category":27},"",[37,24],"meta","Autoresearch","main","504a21d53c0c9b485f6e90c43db83000825e1d3c","skills\u002Fautoresearch",308,"https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon","https:\u002F\u002Fwww.aeon.fun\u002F","https:\u002F\u002Fgithub.com\u002Faeonfun","The most autonomous agent framework.","2026-11-15 03:40:18",{"source":8,"skillName":5,"commands":49,"prompt":59,"zipAvailable":60},[50,53,56],{"tool":51,"command":52},"npx","npx skills add aeonfun\u002Faeon --skill autoresearch",{"tool":54,"command":55},"bunx","bunx skills add aeonfun\u002Faeon --skill autoresearch",{"tool":57,"command":58},"pnpm","pnpm dlx skills add aeonfun\u002Faeon --skill autoresearch","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fskills\u002Fautoresearch\u002FSKILL.md",true,[],[63,70,76,82,88,94],{"skillNo":64,"name":65,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":69},"01M7DSM3QWBJ44FKRMDDRAGNET","Warn Mixed Script","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fwarn-mixed-script\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":71,"name":72,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":69},"01M7DSM0496DRYYW5A4NAAHZDX","Warn Hidden CSS","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fwarn-hidden-css\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":69},"01M7DSKWCGDVYZSDJM31YHS2JB","Malicious Injection Uppercase","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fmalicious-injection-uppercase\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":6,"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":69},"01M7DSKRV2RRA1V0HK13KYWJXD","Malicious Injection Homoglyph","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fmalicious-injection-homoglyph\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":69},"01M7DSKMXJKY0167YHVCKF7B49","Malicious Injection Disregard Prior","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fmalicious-injection-disregard-prior\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":69},"01M7DSKGFQ5V2VNSA4K97CVFMB","Benign Greek Math","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fbenign-greek-math\u002FSKILL.md",[],[]]