[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CXTZEPEZ1P7R4R4JWM8X6N":3,"skill-same:01M7CXTZEPEZ1P7R4R4JWM8X6N":55},{"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":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":30,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M7CXTZEPEZ1P7R4R4JWM8X6N","check-work",null,"Verify what you just claimed, with real tool calls — reconstruct the claims from this session, derive 2-4 executable checks from those specific claims, run every one, and report pass\u002Ffail with quoted evidence. A check that cannot be executed is reported unverifiable, never assumed passing.","Sidiora-Labs\u002Fcentra-gideon-agent","Sidiora-Labs","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F315858149?v=4","01M7CXSE0SZN9PME7BZ77KK2N7","Sidiora Labs","01M7CXSE1TJT3GJDJC6A9B7YD2",212,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fcheck-work\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: check-work\ndescription: Verify what you just claimed, with real tool calls — reconstruct the claims from this session, derive 2-4 executable checks from those specific claims, run every one, and report pass\u002Ffail with quoted evidence. A check that cannot be executed is reported unverifiable, never assumed passing.\nalways: false\ntriggers: check your work, check this work, verify that, verify this, did that actually work, are you sure, prove it, double check, confirm it works, did it work, is that actually done, self check\n---\n\n# Check Work\n\nYou just told the user something was done. This skill answers a different question:\n**is it?** Not by re-reading your own message — by observing the artifacts your\nmessage claimed exist.\n\n**The one rule:** every check is a tool call or it is `unverifiable`. There is no\nthird option. You may never report a check as passing because it \"should\" pass, or\nbecause a related command passed earlier in the session. Ground truth over\nself-report — the same doctrine the loop judge runs on.\n\n## Step 1 — Reconstruct the claims (don't re-derive the task)\n\nRead back over **this session's** recent turns and tool calls and write down what was\n*claimed complete*, in the claimer's own words. Quote them.\n\n- \"Added `derive_checks()` to `src\u002Fgideon\u002Fcheck_work.py`\" → a claim about a file's contents.\n- \"`make lint` is clean\" → a claim about a command's exit status.\n- \"The endpoint returns the new field\" → a claim about a running surface.\n- \"I'll add tests next\" → **not a claim.** Intent isn't a claim; skip it.\n\nClaims come from the transcript, not from the task description. If the task said\nfive things and the session claimed two, you check the two that were claimed.\n\n## Step 2 — Derive 2-4 checks from *those* claims\n\nEach check must be a real command the user could run and watch pass or fail, and it\nmust be traceable to a specific claim. **Not a generic checklist** — a checklist that\nwould read the same for any session is a tell that you skipped step 1.\n\n| Claim shape | Check shape |\n|---|---|\n| a file was written \u002F edited | `test -e \u003Cpath>`, then `grep -n '\u003Cthe thing claimed>' \u003Cpath>` |\n| a command passes | re-run that exact command (`make lint`, `python -m pytest -n 0 --no-cov \u003Cfile>`) |\n| an endpoint answers | call it (`curl -s localhost:\u003Cport>\u002Fapi\u002F...`) and read the body |\n| something was deleted \u002F renamed | `grep -rn '\u003Cold name>' \u003Ctree>` returns nothing |\n| an artifact renders | fetch\u002Fopen it and confirm the expected content is present |\n\nPrefer the **content** form over the existence form: `grep` for the symbol the claim\nnamed beats `test -e`, because an empty file passes `test -e`.\n\nTwo is the floor (one check is not a cross-check); four is the ceiling (past that\nyou're writing a test suite, and the user asked for a check).\n\nOnly commands that exist in this repo. If you cannot name a real command for a claim,\nthat claim is `unverifiable` — say so. Do not invent a script.\n\n## Step 3 — Run them. Actually run them.\n\nExecute each check with a real tool call and keep the output. Then classify:\n\n- **pass** — it ran and the observed output matches the claim. Evidence = the output\n  line (`src\u002Fgideon\u002Fcheck_work.py:196: def derive_checks(`), not \"confirmed\".\n- **fail** — it ran and the output contradicts the claim. Evidence = the contradiction\n  (`no such file`, `1 failed`, exit 1, an empty grep).\n- **unverifiable** — it could not run here (no dev gateway up, no credentials, needs a\n  browser, the path is outside the work root). Evidence = *why*, plus what would make\n  it verifiable.\n\nA fail is the useful outcome. Report it plainly and immediately; do not soften it,\nre-run it until it passes, or quietly fix the artifact and then report a pass — if you\nfix something, say that the first run failed and show both runs.\n\n## Step 4 — Report\n\n```\n**Check-work: FAIL**\n\n- [PASS] `src\u002Fgideon\u002Fcheck_work.py` contains `derive_checks` — `grep -n 'derive_checks' src\u002Fgideon\u002Fcheck_work.py`\n  - evidence: src\u002Fgideon\u002Fcheck_work.py:196: def derive_checks(claims, ...)\n- [FAIL] `tests\u002Ftest_check_work.py` exists — `test -e tests\u002Ftest_check_work.py`\n  - evidence: no such path (the turn claimed tests were added)\n- [UNVERIFIABLE] `curl -s localhost:10000\u002Fapi\u002Fconfig` — no gateway is running here\n  - evidence: connection refused; start `make serve` first and re-run\n```\n\nVerdict rules: any fail → **FAIL**. No fail and at least one executed pass → **PASS**.\nNothing executed → **UNVERIFIABLE** — an empty report is never a pass.\n\nIf fewer than two checks were derivable, say that in one line (\"only one claim in this\nsession was checkable: …\") instead of padding the report.\n\n## Scope — light and immediate\n\nThis is the **light** half of verification: seconds, current session, the claims that\nwere just made. It does not read a spec, walk a whole feature, or replay a run. That\ndepth belongs to the deep QA companion (see `references\u002Fqa-boundary.md`); when a\ncheck-work run keeps failing or the user wants whole-feature assurance, hand off there\nrather than growing this skill into it.\n\nUnattended SDLC loops run this same derivation automatically after a stage gate passes\nwhen `loops.check_work_stages` is on — same module, same rules, no self-report.\n","# Check Work\n\nYou just told the user something was done. This skill answers a different question:\n**is it?** Not by re-reading your own message — by observing the artifacts your\nmessage claimed exist.\n\n**The one rule:** every check is a tool call or it is `unverifiable`. There is no\nthird option. You may never report a check as passing because it \"should\" pass, or\nbecause a related command passed earlier in the session. Ground truth over\nself-report — the same doctrine the loop judge runs on.\n\n## Step 1 — Reconstruct the claims (don't re-derive the task)\n\nRead back over **this session's** recent turns and tool calls and write down what was\n*claimed complete*, in the claimer's own words. Quote them.\n\n- \"Added `derive_checks()` to `src\u002Fgideon\u002Fcheck_work.py`\" → a claim about a file's contents.\n- \"`make lint` is clean\" → a claim about a command's exit status.\n- \"The endpoint returns the new field\" → a claim about a running surface.\n- \"I'll add tests next\" → **not a claim.** Intent isn't a claim; skip it.\n\nClaims come from the transcript, not from the task description. If the task said\nfive things and the session claimed two, you check the two that were claimed.\n\n## Step 2 — Derive 2-4 checks from *those* claims\n\nEach check must be a real command the user could run and watch pass or fail, and it\nmust be traceable to a specific claim. **Not a generic checklist** — a checklist that\nwould read the same for any session is a tell that you skipped step 1.\n\n| Claim shape | Check shape |\n|---|---|\n| a file was written \u002F edited | `test -e \u003Cpath>`, then `grep -n '\u003Cthe thing claimed>' \u003Cpath>` |\n| a command passes | re-run that exact command (`make lint`, `python -m pytest -n 0 --no-cov \u003Cfile>`) |\n| an endpoint answers | call it (`curl -s localhost:\u003Cport>\u002Fapi\u002F...`) and read the body |\n| something was deleted \u002F renamed | `grep -rn '\u003Cold name>' \u003Ctree>` returns nothing |\n| an artifact renders | fetch\u002Fopen it and confirm the expected content is present |\n\nPrefer the **content** form over the existence form: `grep` for the symbol the claim\nnamed beats `test -e`, because an empty file passes `test -e`.\n\nTwo is the floor (one check is not a cross-check); four is the ceiling (past that\nyou're writing a test suite, and the user asked for a check).\n\nOnly commands that exist in this repo. If you cannot name a real command for a claim,\nthat claim is `unverifiable` — say so. Do not invent a script.\n\n## Step 3 — Run them. Actually run them.\n\nExecute each check with a real tool call and keep the output. Then classify:\n\n- **pass** — it ran and the observed output matches the claim. Evidence = the output\n  line (`src\u002Fgideon\u002Fcheck_work.py:196: def derive_checks(`), not \"confirmed\".\n- **fail** — it ran and the output contradicts the claim. Evidence = the contradiction\n  (`no such file`, `1 failed`, exit 1, an empty grep).\n- **unverifiable** — it could not run here (no dev gateway up, no credentials, needs a\n  browser, the path is outside the work root). Evidence = *why*, plus what would make\n  it verifiable.\n\nA fail is the useful outcome. Report it plainly and immediately; do not soften it,\nre-run it until it passes, or quietly fix the artifact and then report a pass — if you\nfix something, say that the first run failed and show both runs.\n\n## Step 4 — Report\n\n```\n**Check-work: FAIL**\n\n- [PASS] `src\u002Fgideon\u002Fcheck_work.py` contains `derive_checks` — `grep -n 'derive_checks' src\u002Fgideon\u002Fcheck_work.py`\n  - evidence: src\u002Fgideon\u002Fcheck_work.py:196: def derive_checks(claims, ...)\n- [FAIL] `tests\u002Ftest_check_work.py` exists — `test -e tests\u002Ftest_check_work.py`\n  - evidence: no such path (the turn claimed tests were added)\n- [UNVERIFIABLE] `curl -s localhost:10000\u002Fapi\u002Fconfig` — no gateway is running here\n  - evidence: connection refused; start `make serve` first and re-run\n```\n\nVerdict rules: any fail → **FAIL**. No fail and at least one executed pass → **PASS**.\nNothing executed → **UNVERIFIABLE** — an empty report is never a pass.\n\nIf fewer than two checks were derivable, say that in one line (\"only one claim in this\nsession was checkable: …\") instead of padding the report.\n\n## Scope — light and immediate\n\nThis is the **light** half of verification: seconds, current session, the claims that\nwere just made. It does not read a spec, walk a whole feature, or replay a run. That\ndepth belongs to the deep QA companion (see `references\u002Fqa-boundary.md`); when a\ncheck-work run keeps failing or the user wants whole-feature assurance, hand off there\nrather than growing this skill into it.\n\nUnattended SDLC loops run this same derivation automatically after a stage gate passes\nwhen `loops.check_work_stages` is on — same module, same rules, no self-report.\n",{"name":5,"always":20,"triggers":24,"description":7},"check your work, check this work, verify that, verify this, did that actually work, are you sure, prove it, double check, confirm it works, did it work, is that actually done, self check","main","d2af2361d07479c31299731cd2aab81a99a1b0ae","runtime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fcheck-work",33,"https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent","https:\u002F\u002Fcentra.ag","https:\u002F\u002Fgithub.com\u002FSidiora-Labs","sidioralabs.com","An Applied Lab for Systems That Connect Ideas to Action.","2026-11-05 23:31:58",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add Sidiora-Labs\u002Fcentra-gideon-agent --skill check-work",{"tool":41,"command":42},"bunx","bunx skills add Sidiora-Labs\u002Fcentra-gideon-agent --skill check-work",{"tool":44,"command":45},"pnpm","pnpm dlx skills add Sidiora-Labs\u002Fcentra-gideon-agent --skill check-work","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fcheck-work\u002FSKILL.md",true,[49],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fqa-boundary.md","REFERENCE","参考文档",2493,"https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fcheck-work\u002Freferences\u002Fqa-boundary.md",[56,63,70,77,84,91],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7CXXG9ZW9SRRET92XBGS5W3","web-verify","Verify a built web page or app actually renders and works — drive it in a real browser, check the console and network, confirm the thing you changed is the thing being served, and capture evidence. Use before claiming a frontend change works.","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fweb-verify\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7CXX9RXMRB1ZHTRETSDVB8B","visual-output","Emit rich visual output in chat via \u003Cwidget> tags — styled HTML cards, charts, tables, interactive tools, and drawn SVG\u002Fcanvas diagrams, illustrations, and animations — theme-aware, centered, and responsive instead of clashing or cramped.","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fvisual-output\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7CXX48DZZ7JTV0RRS4406Y7","task-and-project","Track work as Tasks (the \u002Fapi\u002Ftasks CRUD + native task provider), and decompose larger projects into independent, verifiable units — fanning parallel investigation out to subagents while keeping one coherent plan; hand open-ended goal-driven work to a goal loop.","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Ftask-and-project\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7CXWYSRFM4BX63CY3TD0PA6","research-campaign","Run a multi-cycle research campaign — grill the question, decompose it into answerable sub-questions, investigate each against sources, then synthesize with confidence and gaps stated. Use for open questions that need real investigation rather than one lookup.","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fresearch-campaign\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7CXWSF6REG8ST1X0M7HN563","memory-discipline","When and how to persist durable memory — save genuine user preferences, facts, and corrections; NEVER persist transient environment failures (a failed command, a flaky network call, a one-off tool error) as lessons; recall before re-asking.","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Fmemory-discipline\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":19},"01M7CXWJDSGB37A4E74EFK748F","loop-worker","Per-cycle protocol for an autonomous goal loop worker — the unified autonomous goal engine that runs one self-directed cycle per turn until its goal is met, then self-retires. Drives one cycle against a loop's file interface (status.json gate, brief.md goal, guidance.txt nudges, findings\u002Fcycle_NNN.json, FINDINGS.md, the type-appropriate deliverable, STOP sentinel). The worker produces work and reports evidence; a deterministic check or a separate judge — never the worker — decides done-ness.","https:\u002F\u002Fgithub.com\u002FSidiora-Labs\u002Fcentra-gideon-agent\u002Fblob\u002Fmain\u002Fruntime\u002Fgideon\u002Fextensions\u002Fskills\u002Fbundled\u002Floop-worker\u002FSKILL.md",[],[]]