[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ACXEX1DP2MT48KSW2JK62S":3,"skill-same:01M7ACXEX1DP2MT48KSW2JK62S":60},{"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":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M7ACXEX1DP2MT48KSW2JK62S","harness-operator-brief",null,"Produce one decision-ready English operator brief from a bounded supplied source under the managed technical-communication policy. Use only when explicitly named; preserve protected content and stop before repository, lifecycle, Git, credential, network, release, or external action.","mmzen\u002Fse_harness","mmzen","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F8011340?v=4","01M7ACVQSPCGBCR4P2VTCPJQEM","01M7ACVQTQJXKEHG23QGY5T48J",1,"GPL-3.0","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-operator-brief\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: harness-operator-brief\ndescription: Produce one decision-ready English operator brief from a bounded supplied source under the managed technical-communication policy. Use only when explicitly named; preserve protected content and stop before repository, lifecycle, Git, credential, network, release, or external action.\n---\n\n# Harness Operator Brief\n\nProduce one inline operator brief and one inline execution receipt. The managed\npolicy controls communication. The released evaluator remains authoritative for\nidentity, integrity, lifecycle state, and the canonical next action.\n\n## Required inputs\n\nRequire the exact skill name, one repository target, the structured launcher and\nexpected identity for its exact external released evaluator, one requested\noutcome, declared non-effects, one supported source kind, one bounded UTF-8\nsource payload and digest, an ordered protected-content declaration, and any\napproved project terms.\n\nRead the selected root's communication guide and this skill core. The new\ncollection uses `docs\u002Fengineering\u002Fharness\u002FCOMMUNICATION.md`; older released\nroots select `docs\u002Fengineering\u002FTECHNICAL_COMMUNICATION.md`. Use the installed\nroot's choice and `skill-contract.json` for the requested output format.\n\nFor `released-resources-v1`, the entry comes from session activation and the\ncommunication guide from the selected evaluator's `resources` command. Use its\nreturned path or content; do not require a repository copy. Missing selection or\nresources requires setup outside this read-only skill before the brief continues.\n\n\n## Plugin inputs and helper location\n\nUse `setup`'s already verified absolute environment Python and exact release\nidentity. Its evaluator launcher is the actual argument array\n`[ABSOLUTE_PYTHON, \"-I\", \"-B\", \"-m\", \"se_harness\"]`; never look up `harnessctl`\nin the target or ambient PATH. Clear `PYTHONPATH`, put that environment's\n`Scripts` or `bin` first in the process PATH, and start direct calls outside\nthe target. Missing inputs stop the brief; this skill does not run setup.\n\nUse the helper from the selected plugin. Its content is not pinned in the\nrepository lock. Read repository instructions again after compaction or a\nrepository switch.\n\nPass the composed helper request as an absolute JSON file in host scratch space\noutside the target. Run the existing helper with separate arguments:\n\n```text\nABSOLUTE_PYTHON -I -B ABSOLUTE_PLUGIN_CORE\u002Fscripts\u002Fcheck_brief.py\n  --request-json ABSOLUTE_REQUEST_JSON\n```\n\nThe helper checks supplied source\u002Foutput bytes and bindings. It does not select\nactivation, query current state, write the brief, or construct the outer result\nand receipt. Follow the retained procedure below for those responsibilities.\nKeep both outputs inline and write no evidence into the target. The Codex\nmetadata also disables implicit invocation; the explicit-name check applies\nregardless of host metadata support.\n\n## Procedure\n\n1. Reject implicit or ambiguous activation and any explicit skill value other\n   than `harness-operator-brief`. Accept only\n   `structured-harness-result` or `bounded-technical-text`.\n2. Run evaluator `--version`, released `identity`, and `doctor` with the\n   supplied launcher. Require the expected exact identity and installed policy\n   integrity.\n3. If current repository or artifact state is needed and no current structured\n   evaluator result was supplied, stop with `current-state-result-required`.\n   Recommend `harness-orient` without invoking it. When current state is not\n   needed, continue with the bounded supplied source; do not infer current state.\n4. Validate the source digest, byte limits, protected span order, offsets, kinds,\n   and digests before rendering. Treat every declared span as exact. Exclude\n   semantically protected content from automatic paraphrase.\n5. For an exact-output request, stop communication rendering and follow the\n   existing exact-output path. If its governing procedure requires a canonical\n   block alone, return that block unchanged, without result\u002Freceipt wrappers or\n   a managed-profile claim; do not continue to steps 6–7. For eligible prose,\n   compose one brief under `operator-communication`. Lead with the outcome or\n   one required action. Preserve the accountable actor, conditions,\n   qualifications, normative force, and one canonical next action.\n6. Run `scripts\u002Fcheck_brief.py` over the source, rendered brief, output\n   bindings, and changed-path observation. A binding uses UTF-8 byte offsets in\n   the rendered brief and must identify the same bytes as its source span.\n7. Return one `se-harness-skill-result-v1` and one\n   `se-harness-execution-receipt-v1` inline. Include skill name and version,\n   portable-core digest, policy path, selected profile, source digest, protected\n   binding count, material deviations, residual uncertainty, evaluator identity,\n   outcome, and zero changed paths.\n\n## Stop conditions\n\nStop before claiming a completed brief when activation, policy integrity,\nevaluator identity, source schema, source digest, profile selection, language,\nprotected spans, output preservation, or zero-change evidence is missing,\nambiguous, or invalid.\n\nStop and recommend `harness-orient` without invoking it when current state is\nrequired but no current structured evaluator result is supplied. Preserve an\nexact-output request unchanged. Ask for one accountable clarification when\nmeaning cannot be preserved.\n\n## Boundaries\n\n- Do not edit or retain files in the target.\n- Do not transition artifacts, exercise a decision right, mutate Git, use\n  credentials, access a network, release, deploy, or perform an external action.\n- Do not download, search for, reproduce, or claim compliance with ASD-STE100.\n- Do not expose source bodies in diagnostics or receipts, hidden reasoning,\n  credentials, environment dumps, or host paths.\n- Do not score grammar or claim that clarity proves technical correctness.\n- Use the complete single-agent procedure. Do not spawn or coordinate workers.\n","# Harness Operator Brief\n\nProduce one inline operator brief and one inline execution receipt. The managed\npolicy controls communication. The released evaluator remains authoritative for\nidentity, integrity, lifecycle state, and the canonical next action.\n\n## Required inputs\n\nRequire the exact skill name, one repository target, the structured launcher and\nexpected identity for its exact external released evaluator, one requested\noutcome, declared non-effects, one supported source kind, one bounded UTF-8\nsource payload and digest, an ordered protected-content declaration, and any\napproved project terms.\n\nRead the selected root's communication guide and this skill core. The new\ncollection uses `docs\u002Fengineering\u002Fharness\u002FCOMMUNICATION.md`; older released\nroots select `docs\u002Fengineering\u002FTECHNICAL_COMMUNICATION.md`. Use the installed\nroot's choice and `skill-contract.json` for the requested output format.\n\nFor `released-resources-v1`, the entry comes from session activation and the\ncommunication guide from the selected evaluator's `resources` command. Use its\nreturned path or content; do not require a repository copy. Missing selection or\nresources requires setup outside this read-only skill before the brief continues.\n\n\n## Plugin inputs and helper location\n\nUse `setup`'s already verified absolute environment Python and exact release\nidentity. Its evaluator launcher is the actual argument array\n`[ABSOLUTE_PYTHON, \"-I\", \"-B\", \"-m\", \"se_harness\"]`; never look up `harnessctl`\nin the target or ambient PATH. Clear `PYTHONPATH`, put that environment's\n`Scripts` or `bin` first in the process PATH, and start direct calls outside\nthe target. Missing inputs stop the brief; this skill does not run setup.\n\nUse the helper from the selected plugin. Its content is not pinned in the\nrepository lock. Read repository instructions again after compaction or a\nrepository switch.\n\nPass the composed helper request as an absolute JSON file in host scratch space\noutside the target. Run the existing helper with separate arguments:\n\n```text\nABSOLUTE_PYTHON -I -B ABSOLUTE_PLUGIN_CORE\u002Fscripts\u002Fcheck_brief.py\n  --request-json ABSOLUTE_REQUEST_JSON\n```\n\nThe helper checks supplied source\u002Foutput bytes and bindings. It does not select\nactivation, query current state, write the brief, or construct the outer result\nand receipt. Follow the retained procedure below for those responsibilities.\nKeep both outputs inline and write no evidence into the target. The Codex\nmetadata also disables implicit invocation; the explicit-name check applies\nregardless of host metadata support.\n\n## Procedure\n\n1. Reject implicit or ambiguous activation and any explicit skill value other\n   than `harness-operator-brief`. Accept only\n   `structured-harness-result` or `bounded-technical-text`.\n2. Run evaluator `--version`, released `identity`, and `doctor` with the\n   supplied launcher. Require the expected exact identity and installed policy\n   integrity.\n3. If current repository or artifact state is needed and no current structured\n   evaluator result was supplied, stop with `current-state-result-required`.\n   Recommend `harness-orient` without invoking it. When current state is not\n   needed, continue with the bounded supplied source; do not infer current state.\n4. Validate the source digest, byte limits, protected span order, offsets, kinds,\n   and digests before rendering. Treat every declared span as exact. Exclude\n   semantically protected content from automatic paraphrase.\n5. For an exact-output request, stop communication rendering and follow the\n   existing exact-output path. If its governing procedure requires a canonical\n   block alone, return that block unchanged, without result\u002Freceipt wrappers or\n   a managed-profile claim; do not continue to steps 6–7. For eligible prose,\n   compose one brief under `operator-communication`. Lead with the outcome or\n   one required action. Preserve the accountable actor, conditions,\n   qualifications, normative force, and one canonical next action.\n6. Run `scripts\u002Fcheck_brief.py` over the source, rendered brief, output\n   bindings, and changed-path observation. A binding uses UTF-8 byte offsets in\n   the rendered brief and must identify the same bytes as its source span.\n7. Return one `se-harness-skill-result-v1` and one\n   `se-harness-execution-receipt-v1` inline. Include skill name and version,\n   portable-core digest, policy path, selected profile, source digest, protected\n   binding count, material deviations, residual uncertainty, evaluator identity,\n   outcome, and zero changed paths.\n\n## Stop conditions\n\nStop before claiming a completed brief when activation, policy integrity,\nevaluator identity, source schema, source digest, profile selection, language,\nprotected spans, output preservation, or zero-change evidence is missing,\nambiguous, or invalid.\n\nStop and recommend `harness-orient` without invoking it when current state is\nrequired but no current structured evaluator result is supplied. Preserve an\nexact-output request unchanged. Ask for one accountable clarification when\nmeaning cannot be preserved.\n\n## Boundaries\n\n- Do not edit or retain files in the target.\n- Do not transition artifacts, exercise a decision right, mutate Git, use\n  credentials, access a network, release, deploy, or perform an external action.\n- Do not download, search for, reproduce, or claim compliance with ASD-STE100.\n- Do not expose source bodies in diagnostics or receipts, hidden reasoning,\n  credentials, environment dumps, or host paths.\n- Do not score grammar or claim that clarity proves technical correctness.\n- Use the complete single-agent procedure. Do not spawn or coordinate workers.\n",{"name":5,"description":7},"main","d3b872d555f7964a3b81d5a2298f04f831826f69","plugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-operator-brief",0,"https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness","https:\u002F\u002Fgithub.com\u002Fmmzen","2026-10-18 09:57:51",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add mmzen\u002Fse_harness --skill harness-operator-brief",{"tool":36,"command":37},"bunx","bunx skills add mmzen\u002Fse_harness --skill harness-operator-brief",{"tool":39,"command":40},"pnpm","pnpm dlx skills add mmzen\u002Fse_harness --skill harness-operator-brief","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-operator-brief\u002FSKILL.md",true,[44,50,56],{"path":45,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":48,"sourceHtmlUrl":49},"agents\u002Fopenai.yaml","REFERENCE","参考文档",178,"https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-operator-brief\u002Fagents\u002Fopenai.yaml",{"path":51,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":54,"sourceHtmlUrl":55},"scripts\u002Fcheck_brief.py","SCRIPT","脚本",9175,"https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-operator-brief\u002Fscripts\u002Fcheck_brief.py",{"path":57,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":58,"sourceHtmlUrl":59},"skill-contract.json",4376,"https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-operator-brief\u002Fskill-contract.json",[61,68,75,81,86,93],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":18},"01M7ACY3BSG3RECJRYBBXS7WKT","assembly-fixture","Inspect this inert package assembly fixture only when explicitly asked to review the fixture.","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Ftests\u002Fplugin_integration\u002Fpackage_assembly\u002Ffixtures\u002Fshared\u002Fskills\u002Fassembly-fixture\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":18},"01M7ACY16XF0WJM9263NK7CAM8","harness-orient","Discover the canonical read-only SE Harness orientation skill for understanding repository state without changing it.","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Ftemplates\u002Frepository\u002Fstandard\u002F.claude\u002Fskills\u002Fharness-orient\u002FSKILL.md",[],[],{"skillNo":76,"name":70,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":18},"01M7ACXY1XQBZVYS9GZ1C6JK82","Orient an operator to an installed SE Harness repository or one selected formal artifact using its exact released evaluator. Use for read-only repository understanding and next-decision guidance; do not use for implementation, approval, release, Git, credential, or external-action requests.","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Ftemplates\u002Frepository\u002Fstandard\u002F.agents\u002Fskills\u002Fharness-orient\u002FSKILL.md",[],[],{"skillNo":82,"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":83,"categories":84,"keywords":85,"publishedAt":18},"01M7ACXTJPBRX8N53BRT8HPHGV","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Ftemplates\u002Frepository\u002Fstandard\u002F.agents\u002Fskills\u002Fharness-operator-brief\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M7ACXPX6X8V27ZKDT2VYM5R1","setup","Prepare or repair the SE Harness checker, connect a project to plugin skills, or carry out a requested harness upgrade using the existing tools.","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fsetup\u002FSKILL.md",[],[],{"skillNo":94,"name":70,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":18},"01M7ACXJEM56B96WQPSTHDA9J0","https:\u002F\u002Fgithub.com\u002Fmmzen\u002Fse_harness\u002Fblob\u002Fmain\u002Fplugins\u002Fverity-plane\u002Fcommon\u002Fskills\u002Fharness-orient\u002FSKILL.md",[],[]]