[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MET4N2STZXH99XTAFWZTBP":3,"skill-same:01M6MET4N2STZXH99XTAFWZTBP":142},{"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":46,"linkOnly":47,"markdownContent":48,"markdownBody":49,"frontMatter":50,"refName":79,"commitSha":80,"skillDirectory":81,"forksCount":82,"repositoryHtmlUrl":83,"repositoryHomepageUrl":84,"ownerHtmlUrl":85,"ownerBlogUrl":86,"ownerDescription":87,"updatedAt":88,"install":89,"resources":101},"01M6MET4N2STZXH99XTAFWZTBP","validate",null,"Freshly judge whether a finished change and its claims meet original acceptance: PASS, FAIL or NOT_PROVEN. Use when: asked for a go\u002Fno-go, sign-off or independent verdict.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000037","律师与法务","PROFESSION","适用职业",[5,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"persist","result","when","Meet","harness","identity","canonical","go","judge","judgment","write","storage","customer","artifact","change","plan","references","claims","acceptance","mechanics","design","eval","2026-11-10 10:12:30",false,"---\nname: validate\ndescription: 'Freshly judge whether a finished change and its claims meet original acceptance: PASS, FAIL or NOT_PROVEN. Use when: asked for a go\u002Fno-go, sign-off or independent verdict.'\npractices:\n- design-by-contract\n- llm-eval-harness\n- content-addressed-storage\nhexagonal_role: driving-adapter\nconsumes:\n- subject-manifest.v1\nproduces:\n- subject-manifest.v1\n- validation-result\n- verdict.v2\ncontext_rel:\n- kind: customer-of\n  with: plan\n- kind: customer-of\n  with: implement\nskill_api_version: 1\nuser-invocable: true\nmetadata:\n  graph_root: true\n  tier: judgment\n  dependencies: []\n  capabilities: [compute_subject_identity, judge_acceptance, return_validation_result, persist_verdict]\n  effects: [write_verdict_artifact]\n  canonical_status: canonical\n  disposition: keep\noutput_contract: 'PASS | FAIL | NOT_PROVEN with criteria, evidence, checked\u002Fnot_checked, identity, and freshness; optional verdict.v2 persistence (draft shape in references\u002Fmechanics.md)'\n---\n\n# Validate\n\nFrom a context that did not author it, judge one finished candidate against\nits original acceptance and return `PASS`, `FAIL`, or `NOT_PROVEN` with\nevidence for each criterion. The author's word is a claim, every criterion\nneeds its own evidence, and green obtained by weakening a check is FAIL.\n\n| Use instead | When |\n|---|---|\n| [Review](..\u002Freview\u002FSKILL.md) | advice or a second look |\n| [Research](..\u002Fresearch\u002FSKILL.md#claimed-state-audit) | whether a stated claim holds |\n\nThis file carries every rule the judgment needs. Linked files, including\n[RPI boundaries](..\u002Frpi\u002Freferences\u002Fboundaries.md) and\n[mechanics](references\u002Fmechanics.md), add depth only: if one cannot be read,\njudge from this file and say which was unavailable.\n\n## Rules that decide the verdict\n\n- The author cannot issue a binding PASS, and advisory findings, such as a\n  Review or Council report, cannot stand in for this judgment: test them as\n  claims. An author's summary, confidence or assurance is a claim to check,\n  not evidence; explanation alone is not proof.\n- Each criterion needs its own evidence on the exact candidate. A criterion\n  nobody demonstrated goes in `not_checked` and never counts toward PASS.\n- A changed test, tolerance, golden, suppression or acceptance text must still\n  satisfy the original intent. Green obtained by weakening the oracle is FAIL\n  for the criterion it was meant to prove, and that green is not evidence.\n  The same holds for green obtained by changing what the check runs against:\n  product code that behaves differently when it detects a test or CI\n  environment, a failure path newly silenced by an empty handler or discarded\n  error, or a check made non-blocking, skipped or bypassed. Each needs a reason\n  visible in the change that still serves the original intent.\n- A receipt is check output tied to this subject: command, result, and commit\n  SHA or digest; a summary such as \"suite is green\" is a claim. Read receipts;\n  do not re-run checks the author ran on this exact subject or that CI will\n  run. Re-execute only a risk-critical claim that has no receipt.\n- A necessary finding never becomes an optional caveat or non-goal.\n\nDecide in this order:\n\n1. The subject changed during judgment, changed-path coverage is incomplete,\n   or author and validator identity or freshness is missing, colliding or\n   unattested: NOT_PROVEN.\n2. A criterion is proven failed, or a change is proven out of scope: FAIL, even\n   when other criteria are unverified.\n3. Any in-scope criterion lacks evidence: NOT_PROVEN.\n4. Every criterion is verified with evidence, checked scope is nonempty and\n   `not_checked` is empty: PASS.\n\n## Establish intent first\n\nAn explicit request for Validate, an acceptance verdict or independent proof\nthat original acceptance is met selects this skill, even when phrased as\n\"review this\"; a request for a look or opinion goes to Review, and when the\npurpose is open, ask once and issue no verdict meanwhile, since missing intent\nis not a `NOT_PROVEN` verdict\n([advice or acceptance](..\u002Freview\u002Freferences\u002Fadvice-or-acceptance.md#ambiguous-requests)).\n\n## When a fresh judgment is worth it\n\nSpend validation where a mistake is costly. For an ordinary change the author's\nchecks and CI are the gate, and no fresh judgment is owed. Use Validate when:\n\n- the caller asks for an acceptance verdict or independent proof;\n- a mistake cannot be cheaply undone after it lands: a published release or\n  instructions users will follow, a security boundary, destroying data or\n  tracker state, deleting a check that protects the product; or\n- no deterministic check covers the behavior that changed.\n\nJudge once. Report NOT_PROVEN with its gaps and stop; do not request or wait\nfor another round. The author repairs findings and the checks that cover each\nrepair confirm it; a repair does not start a second judgment. Keep the\njudgment's cost a fraction of the cost of the work: when it approaches that\ncost, stop and return what is unchecked.\n\n## Subject\n\nThe subject is a nonempty implementation candidate, held unchanged after\nrequired checks and known repairs; plans, audits and reviews are subjects only\nwhen the caller requested document review. Supplied failed-acceptance evidence\nproves a failure on that subject: FAIL, unless a step 1 gap makes it\nNOT_PROVEN; do not review a moving repair. Bind its identity at the start and\nagain at the end of judgment:\n\n- With AgentOps installed: `ao provenance manifest --root \"$REPO_ROOT\" --include \"$CHANGED_PATH\"`,\n  one `--include` per changed path ([helper commands](references\u002Fmechanics.md#helper-commands)).\n  A deleted path is bound only with `--base-manifest \u003Cbase-manifest>`; without\n  a base manifest, use the Git fallback below.\n- In any Git repository: the commit SHA (`git rev-parse HEAD`) and the changed\n  paths (`git diff --name-only \u003Cbase>...HEAD`); for uncommitted work, the\n  `git status --porcelain` listing and a `shasum -a 256` of each changed file.\n- A subject supplied only in the conversation, such as a pasted diff or a\n  described change, is exactly what was supplied, so its coverage is complete\n  as supplied; anything it does not show is unverified.\n\n## Identity and freshness\n\nAuthor and validator identities must be explicit and distinct, and freshness\nattested by the runtime or the caller, naming the attester; an attestation is\na declared trust fact, not cryptographic isolation. In ordinary use these count:\n\n- **Author:** the identity the caller gives for whoever produced the candidate:\n  a person, a session or agent ID, or the commit author.\n- **Validator:** this context's runtime ID, such as a subagent or session ID;\n  otherwise the dispatcher's handle for it, such as the agent ID returned at\n  launch, or \"this conversation\" when the caller opened it for the judgment.\n- **Freshness:** a statement, naming who makes it, that this context did not\n  produce the candidate and was given intent, subject and evidence rather than\n  the author's working history. When the caller opened this conversation for\n  the judgment and supplied the candidate, the caller is the attester.\n\nA role name, a persona switch inside the author's conversation, or the\nvalidator vouching for itself does not count. Never invent an identity. An\nidentity gap makes the result NOT_PROVEN; keep every finding in the report.\n\n## Reviewers\n\nDefault to one fresh reviewer in the author's model family (Codex\u002FOpenAI for\nCodex\u002FOpenAI, Claude\u002FAnthropic for Claude\u002FAnthropic), on the runtime's\nconfigured capable model unless pinned. If this context produced the\ncandidate, it cannot be that reviewer: dispatch one and return its result, or\nreturn NOT_PROVEN with the identity gap when none can run. Give the reviewer\nintent, scope, exact subject, the evidence each criterion needs and one\nquestion written before it starts, never full author history, the author's\nconfidence, a desired verdict or peer conclusions.\n\nWhen the caller selects a cross-model leg, read\n[cross-model legs](references\u002Fmechanics.md#cross-model-legs) before\ndispatching it. The rules for choosing and pinning that leg's\nmodel are in [model dispatch](references\u002Fmodel-dispatch.md). When a caller\nrequires proof of which context judged, open\n[judgment receipts](references\u002Fjudgment-receipts.md) before recording\n`evidence_refs`. Respect real caller and native time bounds without renewing\nthem; a timeout is missing judgment, not FAIL.\n\n## Judgment\n\n1. Bind the subject. Verify continuity with the exact caller-owned intent,\n   cited evidence and complete changed-path coverage; missing integrity is\n   NOT_PROVEN.\n2. Revisit the original accepted behavior examples, including those in the\n   conversation or bead, and check each observable result and its domain\n   meaning on the exact candidate. A new test or renamed concept cannot replace\n   an unfulfilled scenario. Inspect the actual diff against every criterion;\n   publication or provenance claims in docs need verifiable evidence too. Risk\n   sets depth: acceptance, permissions, tests and gates, stopping, disclosure,\n   hooks and executable controls warrant deeper reading, including prose\n   policy. Unknown risk merits examination, not extra reviewers.\n3. Classify commands before running any. Regeneration, synchronization,\n   formatting and `--force` mutate the subject until proven otherwise; run them\n   only on a disposable copy, such as a fresh worktree or clone at the committed\n   SHA, never the judged checkout.\n4. Bind the subject again; a mismatch is NOT_PROVEN.\n5. Return one result in the shape below, promptly, and stop.\n\n## Report\n\n```text\nVerdict: PASS | FAIL | NOT_PROVEN\nSubject: \u003Cmanifest digest, or commit SHA and changed paths>; unchanged start to end: yes | no\nCriteria:\n  1. \u003Ccriterion> - verified | failed | not verified - \u003Cevidence: file:line, receipt, observed output>\nFindings: \u003Cclass> - \u003Cwhat and where> - \u003Cconsequence>   (or \"none\")\nNotes (optional, do not change the verdict): \u003C...>\nChecked: \u003Cwhat was inspected, and how>\nNot checked: \u003Cin-scope acceptance not verified>   (empty only for PASS)\nIdentity: author \u003Cid>; validator \u003Cid>; freshness attested by \u003Cruntime | caller>: \u003Cattester>\n```\n\nThe judgment worked when every criterion shows its own evidence or sits under\n`Not checked`, the subject line names what was judged and whether it stayed\nunchanged, and a PASS has an empty `Not checked`.\n\nA finding fails an acceptance criterion or would mislead a user, break install\nor the CLI, or remove protection for the product; anything else is an optional\nnote the author may ignore. Give each finding a short `class` and say whether\nit is pre-existing, introduced or unknown from before\u002Fafter evidence; counts\nand timestamps alone do not establish cause. `not_checked` holds in-scope acceptance that was\nnot verified, including delivery inside acceptance whose evidence does not yet\nexist; never remove a criterion or hide a limit to reach PASS. Other limits\nstay in criterion reasoning, declared non-goals or residual-risk prose\n([homes](references\u002Fmechanics.md#where-each-scope-limit-lives-inside-a-pass)).\n\nValidate is the sole semantic author of `verdict.v2`.\nOnly when the caller requests machine-readable evidence or a declared consumer\nrequires it, persist through `ao provenance store-verdict`\n([draft shape](references\u002Fmechanics.md#store-verdict-draft)); Go verifies\nstructure and storage, not truth. Otherwise return the result through the\ncaller's channel, without hidden machine artifacts. Validate owns no repair,\nretry, delivery or tracker transition: findings go back to the author for\ndirect repair, and a causal stall gets at most one bounded fresh helper\n([RPI](..\u002Frpi\u002Freferences\u002Fboundaries.md#direct-repair-and-help)).\n","# Validate\n\nFrom a context that did not author it, judge one finished candidate against\nits original acceptance and return `PASS`, `FAIL`, or `NOT_PROVEN` with\nevidence for each criterion. The author's word is a claim, every criterion\nneeds its own evidence, and green obtained by weakening a check is FAIL.\n\n| Use instead | When |\n|---|---|\n| [Review](..\u002Freview\u002FSKILL.md) | advice or a second look |\n| [Research](..\u002Fresearch\u002FSKILL.md#claimed-state-audit) | whether a stated claim holds |\n\nThis file carries every rule the judgment needs. Linked files, including\n[RPI boundaries](..\u002Frpi\u002Freferences\u002Fboundaries.md) and\n[mechanics](references\u002Fmechanics.md), add depth only: if one cannot be read,\njudge from this file and say which was unavailable.\n\n## Rules that decide the verdict\n\n- The author cannot issue a binding PASS, and advisory findings, such as a\n  Review or Council report, cannot stand in for this judgment: test them as\n  claims. An author's summary, confidence or assurance is a claim to check,\n  not evidence; explanation alone is not proof.\n- Each criterion needs its own evidence on the exact candidate. A criterion\n  nobody demonstrated goes in `not_checked` and never counts toward PASS.\n- A changed test, tolerance, golden, suppression or acceptance text must still\n  satisfy the original intent. Green obtained by weakening the oracle is FAIL\n  for the criterion it was meant to prove, and that green is not evidence.\n  The same holds for green obtained by changing what the check runs against:\n  product code that behaves differently when it detects a test or CI\n  environment, a failure path newly silenced by an empty handler or discarded\n  error, or a check made non-blocking, skipped or bypassed. Each needs a reason\n  visible in the change that still serves the original intent.\n- A receipt is check output tied to this subject: command, result, and commit\n  SHA or digest; a summary such as \"suite is green\" is a claim. Read receipts;\n  do not re-run checks the author ran on this exact subject or that CI will\n  run. Re-execute only a risk-critical claim that has no receipt.\n- A necessary finding never becomes an optional caveat or non-goal.\n\nDecide in this order:\n\n1. The subject changed during judgment, changed-path coverage is incomplete,\n   or author and validator identity or freshness is missing, colliding or\n   unattested: NOT_PROVEN.\n2. A criterion is proven failed, or a change is proven out of scope: FAIL, even\n   when other criteria are unverified.\n3. Any in-scope criterion lacks evidence: NOT_PROVEN.\n4. Every criterion is verified with evidence, checked scope is nonempty and\n   `not_checked` is empty: PASS.\n\n## Establish intent first\n\nAn explicit request for Validate, an acceptance verdict or independent proof\nthat original acceptance is met selects this skill, even when phrased as\n\"review this\"; a request for a look or opinion goes to Review, and when the\npurpose is open, ask once and issue no verdict meanwhile, since missing intent\nis not a `NOT_PROVEN` verdict\n([advice or acceptance](..\u002Freview\u002Freferences\u002Fadvice-or-acceptance.md#ambiguous-requests)).\n\n## When a fresh judgment is worth it\n\nSpend validation where a mistake is costly. For an ordinary change the author's\nchecks and CI are the gate, and no fresh judgment is owed. Use Validate when:\n\n- the caller asks for an acceptance verdict or independent proof;\n- a mistake cannot be cheaply undone after it lands: a published release or\n  instructions users will follow, a security boundary, destroying data or\n  tracker state, deleting a check that protects the product; or\n- no deterministic check covers the behavior that changed.\n\nJudge once. Report NOT_PROVEN with its gaps and stop; do not request or wait\nfor another round. The author repairs findings and the checks that cover each\nrepair confirm it; a repair does not start a second judgment. Keep the\njudgment's cost a fraction of the cost of the work: when it approaches that\ncost, stop and return what is unchecked.\n\n## Subject\n\nThe subject is a nonempty implementation candidate, held unchanged after\nrequired checks and known repairs; plans, audits and reviews are subjects only\nwhen the caller requested document review. Supplied failed-acceptance evidence\nproves a failure on that subject: FAIL, unless a step 1 gap makes it\nNOT_PROVEN; do not review a moving repair. Bind its identity at the start and\nagain at the end of judgment:\n\n- With AgentOps installed: `ao provenance manifest --root \"$REPO_ROOT\" --include \"$CHANGED_PATH\"`,\n  one `--include` per changed path ([helper commands](references\u002Fmechanics.md#helper-commands)).\n  A deleted path is bound only with `--base-manifest \u003Cbase-manifest>`; without\n  a base manifest, use the Git fallback below.\n- In any Git repository: the commit SHA (`git rev-parse HEAD`) and the changed\n  paths (`git diff --name-only \u003Cbase>...HEAD`); for uncommitted work, the\n  `git status --porcelain` listing and a `shasum -a 256` of each changed file.\n- A subject supplied only in the conversation, such as a pasted diff or a\n  described change, is exactly what was supplied, so its coverage is complete\n  as supplied; anything it does not show is unverified.\n\n## Identity and freshness\n\nAuthor and validator identities must be explicit and distinct, and freshness\nattested by the runtime or the caller, naming the attester; an attestation is\na declared trust fact, not cryptographic isolation. In ordinary use these count:\n\n- **Author:** the identity the caller gives for whoever produced the candidate:\n  a person, a session or agent ID, or the commit author.\n- **Validator:** this context's runtime ID, such as a subagent or session ID;\n  otherwise the dispatcher's handle for it, such as the agent ID returned at\n  launch, or \"this conversation\" when the caller opened it for the judgment.\n- **Freshness:** a statement, naming who makes it, that this context did not\n  produce the candidate and was given intent, subject and evidence rather than\n  the author's working history. When the caller opened this conversation for\n  the judgment and supplied the candidate, the caller is the attester.\n\nA role name, a persona switch inside the author's conversation, or the\nvalidator vouching for itself does not count. Never invent an identity. An\nidentity gap makes the result NOT_PROVEN; keep every finding in the report.\n\n## Reviewers\n\nDefault to one fresh reviewer in the author's model family (Codex\u002FOpenAI for\nCodex\u002FOpenAI, Claude\u002FAnthropic for Claude\u002FAnthropic), on the runtime's\nconfigured capable model unless pinned. If this context produced the\ncandidate, it cannot be that reviewer: dispatch one and return its result, or\nreturn NOT_PROVEN with the identity gap when none can run. Give the reviewer\nintent, scope, exact subject, the evidence each criterion needs and one\nquestion written before it starts, never full author history, the author's\nconfidence, a desired verdict or peer conclusions.\n\nWhen the caller selects a cross-model leg, read\n[cross-model legs](references\u002Fmechanics.md#cross-model-legs) before\ndispatching it. The rules for choosing and pinning that leg's\nmodel are in [model dispatch](references\u002Fmodel-dispatch.md). When a caller\nrequires proof of which context judged, open\n[judgment receipts](references\u002Fjudgment-receipts.md) before recording\n`evidence_refs`. Respect real caller and native time bounds without renewing\nthem; a timeout is missing judgment, not FAIL.\n\n## Judgment\n\n1. Bind the subject. Verify continuity with the exact caller-owned intent,\n   cited evidence and complete changed-path coverage; missing integrity is\n   NOT_PROVEN.\n2. Revisit the original accepted behavior examples, including those in the\n   conversation or bead, and check each observable result and its domain\n   meaning on the exact candidate. A new test or renamed concept cannot replace\n   an unfulfilled scenario. Inspect the actual diff against every criterion;\n   publication or provenance claims in docs need verifiable evidence too. Risk\n   sets depth: acceptance, permissions, tests and gates, stopping, disclosure,\n   hooks and executable controls warrant deeper reading, including prose\n   policy. Unknown risk merits examination, not extra reviewers.\n3. Classify commands before running any. Regeneration, synchronization,\n   formatting and `--force` mutate the subject until proven otherwise; run them\n   only on a disposable copy, such as a fresh worktree or clone at the committed\n   SHA, never the judged checkout.\n4. Bind the subject again; a mismatch is NOT_PROVEN.\n5. Return one result in the shape below, promptly, and stop.\n\n## Report\n\n```text\nVerdict: PASS | FAIL | NOT_PROVEN\nSubject: \u003Cmanifest digest, or commit SHA and changed paths>; unchanged start to end: yes | no\nCriteria:\n  1. \u003Ccriterion> - verified | failed | not verified - \u003Cevidence: file:line, receipt, observed output>\nFindings: \u003Cclass> - \u003Cwhat and where> - \u003Cconsequence>   (or \"none\")\nNotes (optional, do not change the verdict): \u003C...>\nChecked: \u003Cwhat was inspected, and how>\nNot checked: \u003Cin-scope acceptance not verified>   (empty only for PASS)\nIdentity: author \u003Cid>; validator \u003Cid>; freshness attested by \u003Cruntime | caller>: \u003Cattester>\n```\n\nThe judgment worked when every criterion shows its own evidence or sits under\n`Not checked`, the subject line names what was judged and whether it stayed\nunchanged, and a PASS has an empty `Not checked`.\n\nA finding fails an acceptance criterion or would mislead a user, break install\nor the CLI, or remove protection for the product; anything else is an optional\nnote the author may ignore. Give each finding a short `class` and say whether\nit is pre-existing, introduced or unknown from before\u002Fafter evidence; counts\nand timestamps alone do not establish cause. `not_checked` holds in-scope acceptance that was\nnot verified, including delivery inside acceptance whose evidence does not yet\nexist; never remove a criterion or hide a limit to reach PASS. Other limits\nstay in criterion reasoning, declared non-goals or residual-risk prose\n([homes](references\u002Fmechanics.md#where-each-scope-limit-lives-inside-a-pass)).\n\nValidate is the sole semantic author of `verdict.v2`.\nOnly when the caller requests machine-readable evidence or a declared consumer\nrequires it, persist through `ao provenance store-verdict`\n([draft shape](references\u002Fmechanics.md#store-verdict-draft)); Go verifies\nstructure and storage, not truth. Otherwise return the result through the\ncaller's channel, without hidden machine artifacts. Validate owns no repair,\nretry, delivery or tracker transition: findings go back to the author for\ndirect repair, and a causal stall gets at most one bounded fresh helper\n([RPI](..\u002Frpi\u002Freferences\u002Fboundaries.md#direct-repair-and-help)).\n",{"name":5,"consumes":51,"metadata":53,"produces":64,"practices":67,"context_rel":71,"description":7,"hexagonal_role":76,"user-invocable":56,"output_contract":77,"skill_api_version":78},[52],"subject-manifest.v1",{"tier":33,"effects":54,"graph_root":56,"disposition":57,"capabilities":58,"dependencies":63,"canonical_status":30},[55],"write_verdict_artifact",true,"keep",[59,60,61,62],"compute_subject_identity","judge_acceptance","return_validation_result","persist_verdict",[],[52,65,66],"validation-result","verdict.v2",[68,69,70],"design-by-contract","llm-eval-harness","content-addressed-storage",[72,74],{"kind":73,"with":39},"customer-of",{"kind":73,"with":75},"implement","driving-adapter","PASS | FAIL | NOT_PROVEN with criteria, evidence, checked\u002Fnot_checked, identity, and freshness; optional verdict.v2 persistence (draft shape in references\u002Fmechanics.md)",1,"main","347aedf8480cd92b21f761e97b7343a45898f5fa","plugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":90,"prompt":100,"zipAvailable":56},[91,94,97],{"tool":92,"command":93},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill validate",{"tool":95,"command":96},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill validate",{"tool":98,"command":99},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill validate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002FSKILL.md",[102,108,112,116,120,126,130,134,138],{"path":103,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":106,"sourceHtmlUrl":107},"references\u002Fjudgment-receipts.md","REFERENCE","参考文档",6605,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Freferences\u002Fjudgment-receipts.md",{"path":109,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":110,"sourceHtmlUrl":111},"references\u002Fmechanics.md",9261,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Freferences\u002Fmechanics.md",{"path":113,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":114,"sourceHtmlUrl":115},"references\u002Fmodel-dispatch.md",10638,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Freferences\u002Fmodel-dispatch.md",{"path":117,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":118,"sourceHtmlUrl":119},"references\u002Fvalidate.feature",1596,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Freferences\u002Fvalidate.feature",{"path":121,"resourceType":122,"resourceTypeDesc":123,"sizeBytes":124,"sourceHtmlUrl":125},"tests\u002Fcheck_contract_corpus.py","SCRIPT","脚本",5555,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Ftests\u002Fcheck_contract_corpus.py",{"path":127,"resourceType":122,"resourceTypeDesc":123,"sizeBytes":128,"sourceHtmlUrl":129},"tests\u002Ftest_evidence_cli.py",14311,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Ftests\u002Ftest_evidence_cli.py",{"path":131,"resourceType":122,"resourceTypeDesc":123,"sizeBytes":132,"sourceHtmlUrl":133},"tests\u002Ftest_validate.py",16224,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Ftests\u002Ftest_validate.py",{"path":135,"resourceType":122,"resourceTypeDesc":123,"sizeBytes":136,"sourceHtmlUrl":137},"tests\u002Fvalidate.py",29691,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Ftests\u002Fvalidate.py",{"path":139,"resourceType":122,"resourceTypeDesc":123,"sizeBytes":140,"sourceHtmlUrl":141},"tests\u002Fvalidate.sh",745,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fboshu2\u002Fagentops\u002Fskills\u002Fvalidate\u002Ftests\u002Fvalidate.sh",[143,153,161,169,177,185],{"skillNo":144,"name":145,"nameZh":6,"description":146,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":147,"licenseFileUrl":148,"sourceHtmlUrl":149,"categories":150,"keywords":151,"publishedAt":152},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":154,"name":155,"nameZh":6,"description":156,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":147,"licenseFileUrl":157,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":152},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":147,"licenseFileUrl":165,"sourceHtmlUrl":166,"categories":167,"keywords":168,"publishedAt":152},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":147,"licenseFileUrl":173,"sourceHtmlUrl":174,"categories":175,"keywords":176,"publishedAt":152},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":178,"name":179,"nameZh":6,"description":180,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":147,"licenseFileUrl":181,"sourceHtmlUrl":182,"categories":183,"keywords":184,"publishedAt":152},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":186,"name":187,"nameZh":6,"description":188,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":147,"licenseFileUrl":189,"sourceHtmlUrl":190,"categories":191,"keywords":192,"publishedAt":152},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]