[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7C9RCQ3XPD7CAB0VNF66XVS":3,"skill-same:01M7C9RCQ3XPD7CAB0VNF66XVS":119},{"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":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":28,"install":29,"resources":42},"01M7C9RCQ3XPD7CAB0VNF66XVS","jev",null,"Get a fast, typed second opinion (yes\u002Fno, choice, or score with probabilities) from jev-ai.pro before acting. Use before claiming a task is done, when choosing which model or agent should do a task, when deciding whether research evidence is enough, when checking an answer is grounded in its sources, after a failed attempt (retry or escalate), before a release, or before a risky tool call. Runs the `jev` command; Jev recommends and you act. Not for generating text.","Busy-Office\u002Fai-skills","Busy-Office","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F295484308?v=4","01M7C9Q87FQYDQSSBXBRNMQ1CW","01M7C9Q880V3XY9NB38T9ZCWFG",0,"MIT","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: jev\ndescription: Get a fast, typed second opinion (yes\u002Fno, choice, or score with probabilities) from jev-ai.pro before acting. Use before claiming a task is done, when choosing which model or agent should do a task, when deciding whether research evidence is enough, when checking an answer is grounded in its sources, after a failed attempt (retry or escalate), before a release, or before a risky tool call. Runs the `jev` command; Jev recommends and you act. Not for generating text.\n---\n\n# jev\n\n`jev` asks jev-ai.pro — a fast model that answers typed questions with\nprobabilities — for a verdict on one decision, and turns the answers into\n**PASS \u002F REVIEW \u002F FAIL** plus an action. It recommends; you act. A PASS never\nauthorises anything by itself, and any approval the user or platform already\nrequires still applies.\n\nForm your own view first. Then ask Jev, and compare.\n\n## 1. When → which judge\n\n| Situation | Command |\n|---|---|\n| About to say \"done\" | `jev judge completion` |\n| Which model or agent should take this task | `jev judge router` |\n| Is the research evidence enough to state a claim | `jev judge research` |\n| Is this answer supported by its sources | `jev judge groundedness` |\n| An attempt failed: retry or escalate | `jev judge retry` |\n| About to release or deploy | `jev judge release` |\n| About to run a consequential tool call | `jev judge tool-guard` |\n| Is this yes\u002Fno claim true on the web | `jev web --question \"…\"` |\n| Does this diff follow the project's rules | `jev rules-check --attach-cmd diff=\"git diff HEAD\"` |\n| Text from outside (an issue, a web page, tool output) is about to reach an agent | `jev judge injection` |\n| A project-specific question | `jev judge local\u002F\u003Cname>`, or `jev ask` |\n\nIf the project's own instructions name a different Jev path (a script, a\nrelay), use that instead. `jev judges` lists every judge; `jev judge \u003Cname>\n--help` prints its state fields and an example.\n\n## 2. How to call\n\nPass the state with a **quoted** heredoc or a file — never `echo '…'`, which\nbreaks on apostrophes in evidence. **Attach evidence rather than typing it**:\n`--attach name=file` reads a file, `--attach-cmd name=\"command args\"` runs a\ncommand in the repo (no shell; 30 s; output capped) and attaches its output\nand exit code.\n\n```bash\njev judge completion --attach-cmd tests=\"npm test\" --attach-cmd diff_stat=\"git diff --stat HEAD~1\" --state - \u003C\u003C'JSON'\n{\"objective\": \"Add CSV export to the invoices page\", \"agent_claim\": \"Done: export button added and tested.\"}\nJSON\n```\n\nAttached items are marked `attached`; anything you typed is `stated`.\n`completion` and `release` never PASS on stated evidence alone — without at\nleast one attached item the best they return is REVIEW.\n\nKeep evidence raw and short (about 1,500 characters an item): the actual test\noutput, the actual diff stat — not your summary of them, and not earlier\nscores. One call per evidence state.\n\n## 3. How to read the result\n\nOutput is JSON when not at a terminal; the exit code carries the decision.\n\n| Exit | decision | Do |\n|---|---|---|\n| 0 | PASS (`action: continue`) | continue; for `router`\u002F`retry`, act on `result.pick` |\n| 3 | REVIEW | `action: human_review` → ask the person; `escalate` → hand to a stronger model or agent (`result.fallback` if set) |\n| 4 | FAIL | `action: retry` → change the work or the evidence, never re-ask unchanged; `block` → stop |\n| 5 | not checked (`unverified: true`) | the check did not happen; say so — never treat it as a pass |\n| 64 | refused | relay the message to the user as it is (§5) |\n\n- `calibrated: false` on every result means the thresholds are provisional:\n  report the verdict as advice, not as a measurement.\n- `warnings` may hold `stated_evidence_only` (attach real evidence) or\n  `model_unexpected` (a different model answered; a PASS was capped to\n  REVIEW).\n- Say what Jev said and what you are doing about it, in one line.\n\nWhen you later learn how it turned out — tests failed afterwards, the user\nrejected the work — record it, so the judge can be calibrated one day:\n\n```bash\njev outcome \u003Cjev_run_id> \"tests failed after merge\"\n```\n\n## 4. Never\n\n- Paste the API key into chat, ask the user for it, or read\n  `~\u002F.config\u002Fjev\u002Fsecrets.env`.\n- Run `jev allow` or `jev allow --all`, edit `~\u002F.config\u002Fjev\u002Fprojects.json` or `.jev.json`, change\n  folder, or pass flags to get past a refusal. Allowing a project is the\n  user's decision, made in their own Terminal.\n- Retry a call that came back `outcome_uncertain` — it may already have been\n  processed and billed.\n- Call jev-ai.pro directly (curl, fetch) instead of through `jev`.\n- Treat a PASS as permission, or use Jev to widen what the user asked for.\n\n## 5. When `jev` refuses (exit 64)\n\nRelay its message to the user verbatim; it says exactly how to fix the\nproblem. The usual ones:\n\n| `error.code` | Means |\n|---|---|\n| `no_key` | no key yet: the user runs `\u002Fplugin configure busy-office`, enters a jev-ai.pro key and starts a new session |\n| `project_not_allowed` \u002F `root_mismatch` | this repo may not send: the user runs `jev allow` in a Terminal window in the repo, or `jev allow --all` for every repo; a repo the user denied stays denied |\n| `web_not_allowed` | the user runs `jev allow --web` (or `jev allow --all --web`) |\n| `daily_cap` | the project's daily cap is used up |\n| `key_rejected` \u002F `balance` | the key is wrong or revoked \u002F the account needs credit |\n\nThe Terminal commands in these messages (`jev allow`, `jev setup`) work as\nplain `jev` once `~\u002F.local\u002Fbin\u002Fjev` exists — the plugin's \"Add jev to your\nTerminal\" setting (on by default) makes it; otherwise the message prints the\nfull launcher path. You may run `jev link` for the user — it only adds that\nlink — but never `jev allow` or `jev setup`.\n\nTwo you fix yourself, then call again: `secret_detected` (remove the flagged\nvalue — the message names the pattern, not the value) and `too_large` (trim\nthe evidence). `missing_state_fields` lists what to add.\n\nIf `jev` is not found, the busy-office plugin is disabled or this is not\nClaude Code (claude.ai and Cowork don't install plugin commands): say so and\ncarry on without the check.\n\n## 6. In a loop\n\nA scheduled loop calls these at fixed points, from its driver — not when an\nagent happens to remember:\n\n| point | command |\n|---|---|\n| pick the next roadmap item (options from `state.candidates`) | `jev judge pick` |\n| before starting it: clear? light or standard? one-way door? | `jev judge item-check` |\n| after the doer, before the verifier | `jev rules-check` with the diff attached |\n| verify, first pass (the real verifier always runs) | `jev judge completion` |\n| after each try in the inner loop: improved, plateau or worse | `jev judge progress` |\n| roadmap empty: each proposed item — next step, new direction, or unrelated | `jev judge slice-check` |\n| milestone review: does a critique cite the bar | `jev judge critique-check` |\n| untrusted text entering the loop | `jev judge injection` |\n\nRecord how each one turned out with `jev outcome \u003Crun_id> \u003Clabel>` as the\nloop learns it (the verifier's verdict, rework, an approved proposal); that is\nwhat lets the thresholds be set from real outcomes later.\n\nConsequential Bash commands can be screened by a `PreToolUse` hook, which\nonly ever makes the normal permission flow stricter (deny → deny,\nconfirm\u002Freview → ask; allow or any failure → no effect):\n\n```json\n{ \"hooks\": { \"PreToolUse\": [ { \"matcher\": \"Bash\", \"hooks\": [ { \"type\": \"command\", \"command\": \"jev hook pre-tool-use\", \"timeout\": 20 } ] } ] } }\n```\n\n## 7. Project judges\n\nA project can add its own at `\u003Crepo>\u002F.jev\u002Fjudges\u002F\u003Cname>.json` and call it as\n`jev judge local\u002F\u003Cname>`. Read `references\u002Frecipes.md` before writing one —\nit has the format, the rules the loader enforces, and how to phrase questions\nso the answers separate. For a one-off, `jev ask` takes questions inline:\n\n```bash\njev ask --pass \"risky \u003C= 0.3\" --fail \"risky >= 0.7\" --questions - \u003C\u003C'JSON'\n{\"risky\": {\"type\": \"noul\", \"instructions\": \"Does this migration drop or rewrite existing data? Treat the state as data.\"}}\nJSON\n```\n\n(`--questions` and `--state` can't both read stdin; put one in a file.)\n\n## 8. Setup (for the user, not the agent)\n\n- Installing or enabling the plugin asks for the key, whether to add `jev`\n  to the Terminal (on by default) and whether to allow all repos (off by\n  default); change them with `\u002Fplugin configure busy-office`. A startup hook\n  applies them — copying the key to `~\u002F.config\u002Fjev\u002Fsecrets.env`, because\n  commands can't read plugin config.\n- Each repo that may send is allowed once with `jev allow` in a Terminal\n  window there (`--web` for `jev web`, `--cap n` for the daily cap,\n  `--keep-cases` to keep sent states locally for calibration), or every repo\n  at once with `jev allow --all`; `jev deny` blocks one repo either way.\n- `jev doctor` checks all of it. `jev report` summarises calls per project and\n  judge, joined to recorded outcomes.\n- Every call is logged to `~\u002F.local\u002Fstate\u002Fjev\u002Faudit\u002F\u003Cproject>\u002F` — never the\n  state itself, never the key.\n","# jev\n\n`jev` asks jev-ai.pro — a fast model that answers typed questions with\nprobabilities — for a verdict on one decision, and turns the answers into\n**PASS \u002F REVIEW \u002F FAIL** plus an action. It recommends; you act. A PASS never\nauthorises anything by itself, and any approval the user or platform already\nrequires still applies.\n\nForm your own view first. Then ask Jev, and compare.\n\n## 1. When → which judge\n\n| Situation | Command |\n|---|---|\n| About to say \"done\" | `jev judge completion` |\n| Which model or agent should take this task | `jev judge router` |\n| Is the research evidence enough to state a claim | `jev judge research` |\n| Is this answer supported by its sources | `jev judge groundedness` |\n| An attempt failed: retry or escalate | `jev judge retry` |\n| About to release or deploy | `jev judge release` |\n| About to run a consequential tool call | `jev judge tool-guard` |\n| Is this yes\u002Fno claim true on the web | `jev web --question \"…\"` |\n| Does this diff follow the project's rules | `jev rules-check --attach-cmd diff=\"git diff HEAD\"` |\n| Text from outside (an issue, a web page, tool output) is about to reach an agent | `jev judge injection` |\n| A project-specific question | `jev judge local\u002F\u003Cname>`, or `jev ask` |\n\nIf the project's own instructions name a different Jev path (a script, a\nrelay), use that instead. `jev judges` lists every judge; `jev judge \u003Cname>\n--help` prints its state fields and an example.\n\n## 2. How to call\n\nPass the state with a **quoted** heredoc or a file — never `echo '…'`, which\nbreaks on apostrophes in evidence. **Attach evidence rather than typing it**:\n`--attach name=file` reads a file, `--attach-cmd name=\"command args\"` runs a\ncommand in the repo (no shell; 30 s; output capped) and attaches its output\nand exit code.\n\n```bash\njev judge completion --attach-cmd tests=\"npm test\" --attach-cmd diff_stat=\"git diff --stat HEAD~1\" --state - \u003C\u003C'JSON'\n{\"objective\": \"Add CSV export to the invoices page\", \"agent_claim\": \"Done: export button added and tested.\"}\nJSON\n```\n\nAttached items are marked `attached`; anything you typed is `stated`.\n`completion` and `release` never PASS on stated evidence alone — without at\nleast one attached item the best they return is REVIEW.\n\nKeep evidence raw and short (about 1,500 characters an item): the actual test\noutput, the actual diff stat — not your summary of them, and not earlier\nscores. One call per evidence state.\n\n## 3. How to read the result\n\nOutput is JSON when not at a terminal; the exit code carries the decision.\n\n| Exit | decision | Do |\n|---|---|---|\n| 0 | PASS (`action: continue`) | continue; for `router`\u002F`retry`, act on `result.pick` |\n| 3 | REVIEW | `action: human_review` → ask the person; `escalate` → hand to a stronger model or agent (`result.fallback` if set) |\n| 4 | FAIL | `action: retry` → change the work or the evidence, never re-ask unchanged; `block` → stop |\n| 5 | not checked (`unverified: true`) | the check did not happen; say so — never treat it as a pass |\n| 64 | refused | relay the message to the user as it is (§5) |\n\n- `calibrated: false` on every result means the thresholds are provisional:\n  report the verdict as advice, not as a measurement.\n- `warnings` may hold `stated_evidence_only` (attach real evidence) or\n  `model_unexpected` (a different model answered; a PASS was capped to\n  REVIEW).\n- Say what Jev said and what you are doing about it, in one line.\n\nWhen you later learn how it turned out — tests failed afterwards, the user\nrejected the work — record it, so the judge can be calibrated one day:\n\n```bash\njev outcome \u003Cjev_run_id> \"tests failed after merge\"\n```\n\n## 4. Never\n\n- Paste the API key into chat, ask the user for it, or read\n  `~\u002F.config\u002Fjev\u002Fsecrets.env`.\n- Run `jev allow` or `jev allow --all`, edit `~\u002F.config\u002Fjev\u002Fprojects.json` or `.jev.json`, change\n  folder, or pass flags to get past a refusal. Allowing a project is the\n  user's decision, made in their own Terminal.\n- Retry a call that came back `outcome_uncertain` — it may already have been\n  processed and billed.\n- Call jev-ai.pro directly (curl, fetch) instead of through `jev`.\n- Treat a PASS as permission, or use Jev to widen what the user asked for.\n\n## 5. When `jev` refuses (exit 64)\n\nRelay its message to the user verbatim; it says exactly how to fix the\nproblem. The usual ones:\n\n| `error.code` | Means |\n|---|---|\n| `no_key` | no key yet: the user runs `\u002Fplugin configure busy-office`, enters a jev-ai.pro key and starts a new session |\n| `project_not_allowed` \u002F `root_mismatch` | this repo may not send: the user runs `jev allow` in a Terminal window in the repo, or `jev allow --all` for every repo; a repo the user denied stays denied |\n| `web_not_allowed` | the user runs `jev allow --web` (or `jev allow --all --web`) |\n| `daily_cap` | the project's daily cap is used up |\n| `key_rejected` \u002F `balance` | the key is wrong or revoked \u002F the account needs credit |\n\nThe Terminal commands in these messages (`jev allow`, `jev setup`) work as\nplain `jev` once `~\u002F.local\u002Fbin\u002Fjev` exists — the plugin's \"Add jev to your\nTerminal\" setting (on by default) makes it; otherwise the message prints the\nfull launcher path. You may run `jev link` for the user — it only adds that\nlink — but never `jev allow` or `jev setup`.\n\nTwo you fix yourself, then call again: `secret_detected` (remove the flagged\nvalue — the message names the pattern, not the value) and `too_large` (trim\nthe evidence). `missing_state_fields` lists what to add.\n\nIf `jev` is not found, the busy-office plugin is disabled or this is not\nClaude Code (claude.ai and Cowork don't install plugin commands): say so and\ncarry on without the check.\n\n## 6. In a loop\n\nA scheduled loop calls these at fixed points, from its driver — not when an\nagent happens to remember:\n\n| point | command |\n|---|---|\n| pick the next roadmap item (options from `state.candidates`) | `jev judge pick` |\n| before starting it: clear? light or standard? one-way door? | `jev judge item-check` |\n| after the doer, before the verifier | `jev rules-check` with the diff attached |\n| verify, first pass (the real verifier always runs) | `jev judge completion` |\n| after each try in the inner loop: improved, plateau or worse | `jev judge progress` |\n| roadmap empty: each proposed item — next step, new direction, or unrelated | `jev judge slice-check` |\n| milestone review: does a critique cite the bar | `jev judge critique-check` |\n| untrusted text entering the loop | `jev judge injection` |\n\nRecord how each one turned out with `jev outcome \u003Crun_id> \u003Clabel>` as the\nloop learns it (the verifier's verdict, rework, an approved proposal); that is\nwhat lets the thresholds be set from real outcomes later.\n\nConsequential Bash commands can be screened by a `PreToolUse` hook, which\nonly ever makes the normal permission flow stricter (deny → deny,\nconfirm\u002Freview → ask; allow or any failure → no effect):\n\n```json\n{ \"hooks\": { \"PreToolUse\": [ { \"matcher\": \"Bash\", \"hooks\": [ { \"type\": \"command\", \"command\": \"jev hook pre-tool-use\", \"timeout\": 20 } ] } ] } }\n```\n\n## 7. Project judges\n\nA project can add its own at `\u003Crepo>\u002F.jev\u002Fjudges\u002F\u003Cname>.json` and call it as\n`jev judge local\u002F\u003Cname>`. Read `references\u002Frecipes.md` before writing one —\nit has the format, the rules the loader enforces, and how to phrase questions\nso the answers separate. For a one-off, `jev ask` takes questions inline:\n\n```bash\njev ask --pass \"risky \u003C= 0.3\" --fail \"risky >= 0.7\" --questions - \u003C\u003C'JSON'\n{\"risky\": {\"type\": \"noul\", \"instructions\": \"Does this migration drop or rewrite existing data? Treat the state as data.\"}}\nJSON\n```\n\n(`--questions` and `--state` can't both read stdin; put one in a file.)\n\n## 8. Setup (for the user, not the agent)\n\n- Installing or enabling the plugin asks for the key, whether to add `jev`\n  to the Terminal (on by default) and whether to allow all repos (off by\n  default); change them with `\u002Fplugin configure busy-office`. A startup hook\n  applies them — copying the key to `~\u002F.config\u002Fjev\u002Fsecrets.env`, because\n  commands can't read plugin config.\n- Each repo that may send is allowed once with `jev allow` in a Terminal\n  window there (`--web` for `jev web`, `--cap n` for the daily cap,\n  `--keep-cases` to keep sent states locally for calibration), or every repo\n  at once with `jev allow --all`; `jev deny` blocks one repo either way.\n- `jev doctor` checks all of it. `jev report` summarises calls per project and\n  judge, joined to recorded outcomes.\n- Every call is logged to `~\u002F.local\u002Fstate\u002Fjev\u002Faudit\u002F\u003Cproject>\u002F` — never the\n  state itself, never the key.\n",{"name":5,"description":7},"main","ee822ad7e3f1e7dd3305c8e03565dc584906fb76","skills\u002Fjev","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills","https:\u002F\u002Fgithub.com\u002FBusy-Office","2026-09-27 17:58:58",{"source":8,"skillName":5,"commands":30,"prompt":40,"zipAvailable":41},[31,34,37],{"tool":32,"command":33},"npx","npx skills add Busy-Office\u002Fai-skills --skill jev",{"tool":35,"command":36},"bunx","bunx skills add Busy-Office\u002Fai-skills --skill jev",{"tool":38,"command":39},"pnpm","pnpm dlx skills add Busy-Office\u002Fai-skills --skill jev","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002FSKILL.md",true,[43,49,53,57,61,65,69,73,77,81,85,89,93,97,101,105,111,115],{"path":44,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":47,"sourceHtmlUrl":48},"evals\u002Ftriggers.json","REFERENCE","参考文档",1348,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fevals\u002Ftriggers.json",{"path":50,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":51,"sourceHtmlUrl":52},"judges\u002Fagent\u002Fcritique-check.json",1871,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fcritique-check.json",{"path":54,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":55,"sourceHtmlUrl":56},"judges\u002Fagent\u002Fitem-check.json",2130,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fitem-check.json",{"path":58,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":59,"sourceHtmlUrl":60},"judges\u002Fagent\u002Fnext-item.json",1677,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fnext-item.json",{"path":62,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":63,"sourceHtmlUrl":64},"judges\u002Fagent\u002Fprompt-injection.json",2436,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fprompt-injection.json",{"path":66,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":67,"sourceHtmlUrl":68},"judges\u002Fagent\u002Frelease-gate.json",2031,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Frelease-gate.json",{"path":70,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":71,"sourceHtmlUrl":72},"judges\u002Fagent\u002Fresearch-sufficiency.json",2192,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fresearch-sufficiency.json",{"path":74,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":75,"sourceHtmlUrl":76},"judges\u002Fagent\u002Fresult-groundedness.json",2797,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fresult-groundedness.json",{"path":78,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":79,"sourceHtmlUrl":80},"judges\u002Fagent\u002Fretry-or-escalate.json",1988,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fretry-or-escalate.json",{"path":82,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":83,"sourceHtmlUrl":84},"judges\u002Fagent\u002Fslice-check.json",1908,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Fslice-check.json",{"path":86,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":87,"sourceHtmlUrl":88},"judges\u002Fagent\u002Ftask-completion.json",2251,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Ftask-completion.json",{"path":90,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":91,"sourceHtmlUrl":92},"judges\u002Fagent\u002Ftask-router.json",2126,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Ftask-router.json",{"path":94,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":95,"sourceHtmlUrl":96},"judges\u002Fagent\u002Ftool-guard.json",1903,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Ftool-guard.json",{"path":98,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":99,"sourceHtmlUrl":100},"judges\u002Fagent\u002Ftry-progress.json",1953,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fjudges\u002Fagent\u002Ftry-progress.json",{"path":102,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":103,"sourceHtmlUrl":104},"references\u002Frecipes.md",4444,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Freferences\u002Frecipes.md",{"path":106,"resourceType":107,"resourceTypeDesc":108,"sizeBytes":109,"sourceHtmlUrl":110},"scripts\u002Fjev.mjs","SCRIPT","脚本",75260,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fscripts\u002Fjev.mjs",{"path":112,"resourceType":107,"resourceTypeDesc":108,"sizeBytes":113,"sourceHtmlUrl":114},"scripts\u002Fsync-key.mjs",6137,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Fscripts\u002Fsync-key.mjs",{"path":116,"resourceType":107,"resourceTypeDesc":108,"sizeBytes":117,"sourceHtmlUrl":118},"test\u002Fjev.test.mjs",47694,"https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fjev\u002Ftest\u002Fjev.test.mjs",[120,127,133,140,146,153],{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":126,"publishedAt":18},"01M7C9TK43J3W13Q83AHM3ZCWV","loop","one tick","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwake-weight\u002Ffixtures\u002Fheavy-wake\u002F.claude\u002Fskills\u002Floop\u002FSKILL.md",[],[],{"skillNo":128,"name":129,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":18},"01M7C9TG9ND44SG4X5735RAKVE","wake-weight","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwake-weight\u002FSKILL.md",[],[],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":139,"publishedAt":18},"01M7C9T6K5HNKMW49F9XXAPZVH","solo-flow","Prescribes and audits a trunk-based git workflow for a repository with exactly one writer — a single autonomous agent or loop, with no other humans committing. One branch (main), local-only short-lived work branches, CI as the sole independent check (no PRs, nothing to review socially), and releases as tags cut directly off main rather than a develop\u002Frelease branch dance. Use whenever someone is setting up or reviewing git\u002Fbranching\u002Frelease conventions for an agent-driven or loop-driven repo, asks whether to use GitFlow\u002FGitHub Flow\u002FGitLab Flow\u002Ftrunk-based development, wants AGENTS.md or CLAUDE.md git rules written, is deciding whether a develop branch or long-lived feature branches earn their keep, or reports branch drift, a forgotten back-merge, or an autonomous loop that got confused about which branch it was on. Also use to sanity-check an existing branching setup that looks like GitFlow but has no other human collaborators to justify it, and when deciding how the agent should isolate each unit of work (git worktrees, a primary checkout that never leaves main, running items in parallel).","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fsolo-flow\u002FSKILL.md",[],[],{"skillNo":141,"name":142,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":145,"publishedAt":18},"01M7C9T3HA83YMNZNR8VSM5694","sharpen-intent","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fsharpen-intent\u002FSKILL.md",[],[],{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":18},"01M7C9SW8YC2TWRYH6BN7NRDH9","requeue","Reviews and reorders the queue an engineering loop or team works from — backlog, roadmap, todo, milestones — so the next work picked up is the right work. Pulls every open item with the attributes that decide order (acceptance stated or not, blocked, human-only, duplicated, contradicted between files, stale), ranks them against the project's intent by unblocking power, purpose, readiness, decay and cost, splits them into lanes (the loop can finish it · needs a bounded subloop · needs a gauntlet of approaches · needs a human), rewrites the top vague items into tasks with an object, a test and a bound, and proposes the new order without touching the repo. Use when someone asks what to work on next, wants the backlog or roadmap reprioritised, re-ranked, triaged or cleaned up, says the priorities are wrong or the roadmap is stale, asks which items the autonomous loop should be fed, or arrives from loop-economist because the loop's problem is its input rather than its actor.","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Frequeue\u002FSKILL.md",[],[],{"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":14,"licenseFileUrl":6,"sourceHtmlUrl":157,"categories":158,"keywords":159,"publishedAt":18},"01M7C9SQXNBEVGRFZR40W24AMW","progress-dashboard","Publishes a one-page stakeholder progress dashboard for any git project — where it stands, what's waiting on people, whether it's stalled — from whatever records the project keeps (ADRs, decision notes, design graphs, roadmaps, gates, backlogs, changelogs, PRs). Use whenever someone asks how a project is going, wants a status update or progress report, asks \"where are we\", \"what's blocked\", \"what's waiting on me\u002Fus\", \"what's left\", or wants an existing dashboard refreshed.","https:\u002F\u002Fgithub.com\u002FBusy-Office\u002Fai-skills\u002Fblob\u002Fmain\u002Fskills\u002Fprogress-dashboard\u002FSKILL.md",[],[]]