[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M76ZPETP7S0ANSA87RYF6KDB":3,"skill-same:01M76ZPETP7S0ANSA87RYF6KDB":49},{"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":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":31,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M76ZPETP7S0ANSA87RYF6KDB","autoresearch-loop",null,"Run an explicit, bounded modify-verify-decide loop toward a measurable metric with approval gates, scoped edits, and rollback proof.","Lord1Egypt\u002Fawesome-skill-forge","Lord1Egypt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99746359?v=4","01M7592YDNT0XK9WJ0T2MQVPDN","Mohamed Mounir Ali AL Keshita","01M75GRFNHC8VZNWEB8DKDE433",3,"MIT","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Fautoresearch-loop\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: autoresearch-loop\ndescription: \"Run an explicit, bounded modify-verify-decide loop toward a measurable metric with approval gates, scoped edits, and rollback proof.\"\nmetadata:\n  version: \"0.2.1\"\n---\n# Autoresearch Loop\n\nUse this skill only when the user explicitly asks for an autoresearch or iterative improvement loop. It is for bounded, measurable optimisation in a version-controlled workspace. It must not start from casual improvement language, and it must not run unattended until the user has approved the goal, commands, scope, rollback strategy, run mode, and iteration cap.\n\nThe goal is measurable. Each iteration makes one atomic change, verifies it, and keeps or discards the result. The loop stops when the goal is met, the approved iteration cap is reached, the user stops it, or a blocker\u002Fsafety gate is hit.\n\n## Core Loop\n\n```text\n1. Confirm approved run contract\n2. Read context + lessons file\n3. Pick ONE hypothesis\n4. Make ONE atomic change inside approved scope\n5. Snapshot\u002Fcommit before verification\n6. Run approved VERIFY command\n7. Run approved GUARD command\n8. Decision: keep \u002F discard \u002F rework\n9. Log the result\n10. Health check and safety gate\n11. Repeat only within approved cap\n```\n\nRead `references\u002Floop-protocol.md` for the full loop spec.\nRead `references\u002Fpivot-protocol.md` for the escalation ladder.\nRead `references\u002Flessons-protocol.md` for cross-run learning.\n\n## Before Starting\n\nConfirm with the user and do not start until the contract is explicit:\n- **Goal** — one sentence describing what you want to achieve\n- **Metric** — what number is measured, direction, baseline, and target\n- **Verify command** — exact command used to measure the metric\n- **Guard command** — exact command that must keep passing\n- **Scope** — files\u002Fdirectories allowed to change, and files\u002Fdirectories that are forbidden\n- **Rollback strategy** — normal branch\u002Fworktree revert, or isolated disposable reset\n- **Run mode** — foreground by default; background\u002Funattended only after explicit approval\n- **Iteration cap** — required for background\u002Funattended runs; recommended for foreground runs\n- **External research policy** — web\u002Fsearch is off by default unless explicitly approved\n- **Data boundary** — do not expose private code, secrets, logs, or proprietary data to external sources\n\nShow the run contract and ask for confirmation. One round minimum. Then start only after the user says go.\n\n## Verify vs Guard\n\n- **Verify** = \"Did the target metric improve?\" — measures progress\n- **Guard** = \"Did anything else break?\" — prevents regressions\n- Guard files are never modified\n- If verify passes but guard fails: rework up to 2 attempts, then discard\n\n## Decision Rules\n\n| Result | Action |\n|--------|--------|\n| Verify pass + Guard pass | Keep. Extract lesson. |\n| Verify pass + Guard fail | Rework within approved scope (max 2 attempts). If still failing, discard. |\n| Verify fail | Discard using approved rollback. |\n| Crash | Stop unless the fix is clearly inside approved scope and non-destructive. |\n| Syntax error | Fix immediately only if caused by the current iteration and inside approved scope. |\n\n## Escalation Ladder\n\nSee `references\u002Fpivot-protocol.md` for full details.\n\n| Trigger | Action |\n|---------|--------|\n| 3 consecutive discards | REFINE — adjust within current strategy |\n| 5 consecutive discards | PIVOT — abandon strategy, try fundamentally different approach |\n| 2 PIVOTs without improvement | Ask before external research unless pre-approved |\n| 3 PIVOTs without improvement | Soft blocker — stop and report to human |\n\nA single successful keep resets all counters.\n\n## Long Run Hygiene\n\n- Every completed experiment must be recorded before the next one starts\n- Re-read original instructions every 10 iterations to prevent context drift\n- Log: one row per iteration (iteration, commit\u002Fsnapshot, metric, delta, status, description)\n- For background runs, send progress at the approved cadence and stop at the cap\n\n## Lessons\n\nExtract structured lessons after:\n- Every kept iteration (what worked and why)\n- Every PIVOT decision (what failed and why)\n- Run completion\n\nStore in `autoresearch-lessons.md` in the working repo root unless the user chose another path. Do not commit this file unless the user explicitly asks. Consult it at the start of each run. Keep about 50 entries, summarising older ones with time decay.\n\n## Safety\n\n- Foreground mode is the default. Background\u002Funattended mode requires explicit approval and an iteration cap.\n- Make changes only inside approved scope.\n- Commit, snapshot, or otherwise record only your own changes before verification.\n- Revert only changes made by the current loop.\n- Never reset unrelated user work.\n- Never modify guard files unless the user explicitly changes the scope contract.\n- Do not run destructive commands, deploy, publish, push, or touch production systems unless explicitly approved for this run.\n- Do not use web search or external sources unless the run contract allows it.\n- Do not paste private code, secrets, logs, customer data, or proprietary data into external services.\n- Stop and report if the metric cannot be measured mechanically.\n","# Autoresearch Loop\n\nUse this skill only when the user explicitly asks for an autoresearch or iterative improvement loop. It is for bounded, measurable optimisation in a version-controlled workspace. It must not start from casual improvement language, and it must not run unattended until the user has approved the goal, commands, scope, rollback strategy, run mode, and iteration cap.\n\nThe goal is measurable. Each iteration makes one atomic change, verifies it, and keeps or discards the result. The loop stops when the goal is met, the approved iteration cap is reached, the user stops it, or a blocker\u002Fsafety gate is hit.\n\n## Core Loop\n\n```text\n1. Confirm approved run contract\n2. Read context + lessons file\n3. Pick ONE hypothesis\n4. Make ONE atomic change inside approved scope\n5. Snapshot\u002Fcommit before verification\n6. Run approved VERIFY command\n7. Run approved GUARD command\n8. Decision: keep \u002F discard \u002F rework\n9. Log the result\n10. Health check and safety gate\n11. Repeat only within approved cap\n```\n\nRead `references\u002Floop-protocol.md` for the full loop spec.\nRead `references\u002Fpivot-protocol.md` for the escalation ladder.\nRead `references\u002Flessons-protocol.md` for cross-run learning.\n\n## Before Starting\n\nConfirm with the user and do not start until the contract is explicit:\n- **Goal** — one sentence describing what you want to achieve\n- **Metric** — what number is measured, direction, baseline, and target\n- **Verify command** — exact command used to measure the metric\n- **Guard command** — exact command that must keep passing\n- **Scope** — files\u002Fdirectories allowed to change, and files\u002Fdirectories that are forbidden\n- **Rollback strategy** — normal branch\u002Fworktree revert, or isolated disposable reset\n- **Run mode** — foreground by default; background\u002Funattended only after explicit approval\n- **Iteration cap** — required for background\u002Funattended runs; recommended for foreground runs\n- **External research policy** — web\u002Fsearch is off by default unless explicitly approved\n- **Data boundary** — do not expose private code, secrets, logs, or proprietary data to external sources\n\nShow the run contract and ask for confirmation. One round minimum. Then start only after the user says go.\n\n## Verify vs Guard\n\n- **Verify** = \"Did the target metric improve?\" — measures progress\n- **Guard** = \"Did anything else break?\" — prevents regressions\n- Guard files are never modified\n- If verify passes but guard fails: rework up to 2 attempts, then discard\n\n## Decision Rules\n\n| Result | Action |\n|--------|--------|\n| Verify pass + Guard pass | Keep. Extract lesson. |\n| Verify pass + Guard fail | Rework within approved scope (max 2 attempts). If still failing, discard. |\n| Verify fail | Discard using approved rollback. |\n| Crash | Stop unless the fix is clearly inside approved scope and non-destructive. |\n| Syntax error | Fix immediately only if caused by the current iteration and inside approved scope. |\n\n## Escalation Ladder\n\nSee `references\u002Fpivot-protocol.md` for full details.\n\n| Trigger | Action |\n|---------|--------|\n| 3 consecutive discards | REFINE — adjust within current strategy |\n| 5 consecutive discards | PIVOT — abandon strategy, try fundamentally different approach |\n| 2 PIVOTs without improvement | Ask before external research unless pre-approved |\n| 3 PIVOTs without improvement | Soft blocker — stop and report to human |\n\nA single successful keep resets all counters.\n\n## Long Run Hygiene\n\n- Every completed experiment must be recorded before the next one starts\n- Re-read original instructions every 10 iterations to prevent context drift\n- Log: one row per iteration (iteration, commit\u002Fsnapshot, metric, delta, status, description)\n- For background runs, send progress at the approved cadence and stop at the cap\n\n## Lessons\n\nExtract structured lessons after:\n- Every kept iteration (what worked and why)\n- Every PIVOT decision (what failed and why)\n- Run completion\n\nStore in `autoresearch-lessons.md` in the working repo root unless the user chose another path. Do not commit this file unless the user explicitly asks. Consult it at the start of each run. Keep about 50 entries, summarising older ones with time decay.\n\n## Safety\n\n- Foreground mode is the default. Background\u002Funattended mode requires explicit approval and an iteration cap.\n- Make changes only inside approved scope.\n- Commit, snapshot, or otherwise record only your own changes before verification.\n- Revert only changes made by the current loop.\n- Never reset unrelated user work.\n- Never modify guard files unless the user explicitly changes the scope contract.\n- Do not run destructive commands, deploy, publish, push, or touch production systems unless explicitly approved for this run.\n- Do not use web search or external sources unless the run contract allows it.\n- Do not paste private code, secrets, logs, customer data, or proprietary data into external services.\n- Stop and report if the metric cannot be measured mechanically.\n",{"name":5,"metadata":24,"description":7},{"version":25},"0.2.1","master","4aa004da461d820f4ea5499e58eb8d196cf99016","community\u002Fclawhub\u002Fa\u002Fautoresearch-loop",0,"https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge","https:\u002F\u002Fskill-explorer-rho.vercel.app","https:\u002F\u002Fgithub.com\u002FLord1Egypt","Unbreakable. Unstoppable. Unmatched⚡.\r\n Speaker of machine tongues 🤖  | World-builder of decentralized dreams 🌐 | Defender of the digital realm🛡️.","2026-06-10 08:21:01",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add Lord1Egypt\u002Fawesome-skill-forge --skill autoresearch-loop",{"tool":41,"command":42},"bunx","bunx skills add Lord1Egypt\u002Fawesome-skill-forge --skill autoresearch-loop",{"tool":44,"command":45},"pnpm","pnpm dlx skills add Lord1Egypt\u002Fawesome-skill-forge --skill autoresearch-loop","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Fautoresearch-loop\u002FSKILL.md",true,[],[50,58,65,72,79,86],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":57},"01M785WKJB99G3STR1NF9NRTJW","youtube-content","YouTube transcripts to summaries, threads, blogs.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fyoutube-content\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":57},"01M785WEFHB0G2S9YA6EPR3H0E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":57},"01M785W98EDN4EFJXJPMQEHCRC","touchdesigner-mcp","Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftouchdesigner-mcp\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":57},"01M785W5S9FP3QBH2V191E025P","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":57},"01M785W2AS08NSSH9YATCV6FJB","teams-meeting-pipeline","Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fteams-meeting-pipeline\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":57},"01M785VXKW4T34BTP1SCHW1R7F","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[]]