[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C25ZSC032YAGMKAJDXNXSB":3,"skill-same:01M6C25ZSC032YAGMKAJDXNXSB":71},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":53,"repositoryHtmlUrl":54,"repositoryHomepageUrl":6,"ownerHtmlUrl":55,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":56,"install":57,"resources":70},"01M6C25ZSC032YAGMKAJDXNXSB","harness-postmortem",null,"Harness-fix meta-skill (slash: \u002Fmeta) — analyze an agent behavior failure (autonomy violation, refusal, premature stopping) and run \u002Fharness to fix the agent, NOT the underlying task. Input: Slack thread URL, pasted conversation, or freeform description.","jleechanorg\u002Fjleechan-skills","jleechanorg","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F221404966?v=4","01M6C1VJ99XG950X1DWPEBCBGX","01M6C1VJDK5YVDNT3362RT5WSB",6,"MIT","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fharness-postmortem\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"postmortem","harness postmortem","harness","input","autonomy","task","slack thread","freeform","URL","thread","harness-fix","meta-skill","conversation","analyze","fix","Meta","quality","behavior","run","slack","2026-11-10 10:12:30",false,"---\nname: harness-postmortem\ndescription: \"Harness-fix meta-skill (slash: \u002Fmeta) — analyze an agent behavior failure (autonomy violation, refusal, premature stopping) and run \u002Fharness to fix the agent, NOT the underlying task. Input: Slack thread URL, pasted conversation, or freeform description.\"\ntype: quality\nexecution_mode: deferred\nscope: user\n---\n\n# harness-postmortem (slash: \u002Fmeta) — Fix the agent, not the task\n\n**Canonical location:** `~\u002F.hermes\u002Fskills\u002Fharness-postmortem\u002FSKILL.md` (or its `~\u002F.hermes_prod` equivalent). This file is the Claude-side mirror pointer. **Collision rule:** if a workspace contains `.claude\u002Fcommands\u002Fmeta.md`, repo-local content takes precedence over this mirror.\n\nThe slash command is `\u002Fmeta` because that's the user's invocation pattern. Both `meta` and `harness-postmortem` resolve to the same skill.\n\n## Inputs\n\n`$ARGUMENTS` — Slack thread URL, pasted conversation, OR freeform description of agent misbehavior.\n\n## Scope guardrail\n\n`\u002Fmeta` is scope-locked to the **agent behavior failure**. It never investigates or continues the underlying task the agent was attempting — it only analyzes and fixes the harness (instructions, skills, tests, rules) that let the bad behavior happen.\n\n## Behavior\n\n1. Detect input shape (URL \u002F paste \u002F description).\n2. Fetch the thread via `mcp__slack__conversations_replies` if URL.\n3. Apply the scope guardrail (NEVER investigate the underlying task).\n4. Run the Phase 0\u002F1 protocol: MAST + ETCLOVG classification → Observe → Isolate → Simulate → Evaluate.\n5. If user passed `--fix` or used a fullrun trigger (`\u002Fa`, `\u002Ffullrun`, `\u002Ffinish`, \"don't stop\", \"iterate until good\"), implement + commit + push.\n6. Otherwise: output the analysis and wait for approval.\n\n## Implementation reference\n\nFull protocol at `~\u002F.hermes\u002Fskills\u002Fharness-postmortem\u002FSKILL.md` (canonical — read this for the complete MAST\u002FETCLOVG taxonomy, phase steps, and anti-patterns). Highlights:\n\n- **Scope guardrail** — top of SKILL.md.\n- **Phase 0** — MAST taxonomy (Cemri et al., arXiv:2503.13657) + ETCLOVG layer model (arXiv:2606.06324) classification.\n- **Phase 1** — Observe → Isolate → Simulate → Evaluate 4-step spine, with 5-Whys inside Simulate as a prompt heuristic.\n- **Phase 2\u002F3 apply + land** — via the `hermes-deploy-pipeline` self-merge flow on the repository that tracks your Hermes harness.\n- **Anti-patterns** — end of SKILL.md.\n- **Prior art** — Refinex-Space harness-fix, HarnessFix paper, MAST, Reflexion, Observe→Isolate→Simulate→Evaluate pattern.\n\n## Examples\n\n- `\u002Fmeta` — auto-detect from last conversation; if none, prompt for input.\n- `\u002Fmeta hermes stopped halfway again` — freeform.\n- `\u002Fmeta https:\u002F\u002F\u003Cworkspace>.slack.com\u002Farchives\u002F\u003Cchannel-id>\u002Fp\u003Cts>` — Slack thread.\n- `\u002Fmeta --fix \u003Cdescription>` — analyze AND apply harness fix without confirmation.\n\n## Output\n\nPer `finish-the-job` Phase 4 — end-state declaration + proof artifact + judgment calls + zero follow-up question.\n\n## Cross-references\n\n- `\u002Fharness` — `~\u002F.claude\u002Fcommands\u002Fharness.md` (the protocol this skill wraps).\n- `harness-postmortem` skill (canonical) — `~\u002F.hermes\u002Fskills\u002Fharness-postmortem\u002FSKILL.md`.\n- `harness-engineering` — `~\u002F.hermes\u002Fskills\u002Fharness-engineering\u002FSKILL.md` (SOUL.md \u002F TOOLS.md rules).\n- `finish-the-job` — `~\u002F.hermes\u002Fskills\u002Ffinish-the-job\u002FSKILL.md` (end-state contract).\n- `hermes-deploy-pipeline` — `~\u002F.hermes\u002Fskills\u002Fhermes-deploy-pipeline\u002FSKILL.md` (landing recipe).\n","# harness-postmortem (slash: \u002Fmeta) — Fix the agent, not the task\n\n**Canonical location:** `~\u002F.hermes\u002Fskills\u002Fharness-postmortem\u002FSKILL.md` (or its `~\u002F.hermes_prod` equivalent). This file is the Claude-side mirror pointer. **Collision rule:** if a workspace contains `.claude\u002Fcommands\u002Fmeta.md`, repo-local content takes precedence over this mirror.\n\nThe slash command is `\u002Fmeta` because that's the user's invocation pattern. Both `meta` and `harness-postmortem` resolve to the same skill.\n\n## Inputs\n\n`$ARGUMENTS` — Slack thread URL, pasted conversation, OR freeform description of agent misbehavior.\n\n## Scope guardrail\n\n`\u002Fmeta` is scope-locked to the **agent behavior failure**. It never investigates or continues the underlying task the agent was attempting — it only analyzes and fixes the harness (instructions, skills, tests, rules) that let the bad behavior happen.\n\n## Behavior\n\n1. Detect input shape (URL \u002F paste \u002F description).\n2. Fetch the thread via `mcp__slack__conversations_replies` if URL.\n3. Apply the scope guardrail (NEVER investigate the underlying task).\n4. Run the Phase 0\u002F1 protocol: MAST + ETCLOVG classification → Observe → Isolate → Simulate → Evaluate.\n5. If user passed `--fix` or used a fullrun trigger (`\u002Fa`, `\u002Ffullrun`, `\u002Ffinish`, \"don't stop\", \"iterate until good\"), implement + commit + push.\n6. Otherwise: output the analysis and wait for approval.\n\n## Implementation reference\n\nFull protocol at `~\u002F.hermes\u002Fskills\u002Fharness-postmortem\u002FSKILL.md` (canonical — read this for the complete MAST\u002FETCLOVG taxonomy, phase steps, and anti-patterns). Highlights:\n\n- **Scope guardrail** — top of SKILL.md.\n- **Phase 0** — MAST taxonomy (Cemri et al., arXiv:2503.13657) + ETCLOVG layer model (arXiv:2606.06324) classification.\n- **Phase 1** — Observe → Isolate → Simulate → Evaluate 4-step spine, with 5-Whys inside Simulate as a prompt heuristic.\n- **Phase 2\u002F3 apply + land** — via the `hermes-deploy-pipeline` self-merge flow on the repository that tracks your Hermes harness.\n- **Anti-patterns** — end of SKILL.md.\n- **Prior art** — Refinex-Space harness-fix, HarnessFix paper, MAST, Reflexion, Observe→Isolate→Simulate→Evaluate pattern.\n\n## Examples\n\n- `\u002Fmeta` — auto-detect from last conversation; if none, prompt for input.\n- `\u002Fmeta hermes stopped halfway again` — freeform.\n- `\u002Fmeta https:\u002F\u002F\u003Cworkspace>.slack.com\u002Farchives\u002F\u003Cchannel-id>\u002Fp\u003Cts>` — Slack thread.\n- `\u002Fmeta --fix \u003Cdescription>` — analyze AND apply harness fix without confirmation.\n\n## Output\n\nPer `finish-the-job` Phase 4 — end-state declaration + proof artifact + judgment calls + zero follow-up question.\n\n## Cross-references\n\n- `\u002Fharness` — `~\u002F.claude\u002Fcommands\u002Fharness.md` (the protocol this skill wraps).\n- `harness-postmortem` skill (canonical) — `~\u002F.hermes\u002Fskills\u002Fharness-postmortem\u002FSKILL.md`.\n- `harness-engineering` — `~\u002F.hermes\u002Fskills\u002Fharness-engineering\u002FSKILL.md` (SOUL.md \u002F TOOLS.md rules).\n- `finish-the-job` — `~\u002F.hermes\u002Fskills\u002Ffinish-the-job\u002FSKILL.md` (end-state contract).\n- `hermes-deploy-pipeline` — `~\u002F.hermes\u002Fskills\u002Fhermes-deploy-pipeline\u002FSKILL.md` (landing recipe).\n",{"name":5,"type":39,"scope":48,"description":7,"execution_mode":49},"user","deferred","main","8144f00ebafc4306264c63239e4b9fa42e48a90e",".claude\u002Fskills\u002Fharness-postmortem",0,"https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills","https:\u002F\u002Fgithub.com\u002Fjleechanorg","2026-11-12 14:45:07",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add jleechanorg\u002Fjleechan-skills --skill harness-postmortem",{"tool":63,"command":64},"bunx","bunx skills add jleechanorg\u002Fjleechan-skills --skill harness-postmortem",{"tool":66,"command":67},"pnpm","pnpm dlx skills add jleechanorg\u002Fjleechan-skills --skill harness-postmortem","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fharness-postmortem\u002FSKILL.md",true,[],[72,89,111,125,138,153],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":43},"01M716Y5MTJ6GYJZ2ATXCJK8G4","instinct","Use when the user invokes \u002Finstinct or \u002Fin, or asks to \"message Instinct\", \"ask Instinct\", \"tell Instinct\", \"check Instinct\", or read replies from the configured peer assistant. iMessage first, WhatsApp Web as fallback.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Finstinct\u002FSKILL.md",[],[74,48,79,80,81,82,83,84,85,86,87,88],"when","replies","check","use","identity","whatsapp","gmail","read","message","ask",{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":100,"publishedAt":43},"01M716Y4HF7FVVMJSA2CCVN5VF","hr","Compatibility alias for history-resume; produce the same fresh-state, provenance-aware resume briefing.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhr\u002FSKILL.md",[95],{"categoryNo":96,"categoryName":97,"taxonomyType":98,"taxonomyTypeDesc":99},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[91,101,102,103,104,105,106,107,108,109,110],"resume","compatibility","provenance","for","history","produce","alias","briefing","state","fresh",{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":118,"publishedAt":43},"01M716Y1FFNFKFJYYKF8HQA6T8","history-resume","Resume briefing for the current repository branch or worktree using fresh state and admitted history sources.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhistory-resume\u002FSKILL.md",[117],{"categoryNo":96,"categoryName":97,"taxonomyType":98,"taxonomyTypeDesc":99},[105,101,119,120,48,121,108,122,104,123,124,109,110],"history resume","branch","analysis","repository","worktree","sources",{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":43},"01M716XWXJ8EPHK4RH9T14TR85","grok","Alias for the canonical grokbot browser conversation workflow; \u002Fgrok talks to the Grok bot the same way \u002Fdot talks to Dot.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fgrok\u002FSKILL.md",[],[127,132,133,134,104,135,107,35,136,137],"dot","bot","workflow","canonical","browser","grokbot",{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":144,"publishedAt":43},"01M70N9Q07DTE0EW5KFXMMMSJ0","codexa-bench","Compare `codexa` and normal `codex` on a fair split of comparable work by default; use `--strict` for duplicate runs of the exact same task.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcodexa-bench\u002FSKILL.md",[],[145,146,147,82,148,104,149,28,150,151,152],"codexa bench","work","split","default","duplicate","compare","normal","codex",{"skillNo":154,"name":155,"nameZh":6,"description":156,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":157,"licenseFileUrl":158,"sourceHtmlUrl":159,"categories":160,"keywords":170,"publishedAt":43},"01M6C24S2X60SP8BBFGAEKPEGW","frontend-design","Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML\u002FCSS layouts, or when styling\u002Fbeautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffrontend-design\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffrontend-design\u002FSKILL.md",[161,164,167],{"categoryNo":162,"categoryName":163,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":165,"categoryName":166,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":168,"categoryName":169,"taxonomyType":98,"taxonomyTypeDesc":99},"01JBSEEDCAT000000000000032","设计师",[171,172,173,82,174,175,176,48,177,178,179,180,181,182,183,184,185,186,39,187,188,189,190,191,192,193,79,194],"frontend","design","frontend design","generic","ui design","react","html","components","code","artifacts","ui","aesthetics","Production","web components","landing","applications","web","css","create","examples","with","react components","web ui","pages"]