[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78B2J9K0J7FS2TA02A1F9QV":3,"skill-same:01M78B2J9K0J7FS2TA02A1F9QV":51},{"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":30,"commitSha":31,"skillDirectory":32,"forksCount":33,"repositoryHtmlUrl":34,"repositoryHomepageUrl":6,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":36,"install":37,"resources":50},"01M78B2J9K0J7FS2TA02A1F9QV","multi-agent-task-decomposition",null,"Decide when a task needs decomposition, define dependency-aware work boundaries, and coordinate sub-agents through explicit contracts and verified integration.","hermes-jarvis-bot\u002Fhermes-agent-config-kit","hermes-jarvis-bot","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F288284026?v=4","01M78AV3BBD4M67J3YWSQRKP1Q","01M78AV3C2SKH53E0PZMXAKTTS",1,"MIT","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fmulti-agent-task-decomposition\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: multi-agent-task-decomposition\ndescription: \"Decide when a task needs decomposition, define dependency-aware work boundaries, and coordinate sub-agents through explicit contracts and verified integration.\"\nversion: 0.1.0\nlicense: MIT\nmetadata:\n  hermes_config_kit:\n    source_repo: AnastasiyaW\u002Fclaude-code-config\n    source_path: principles\u002F06-multi-agent-decomposition.md\n    adapter: hermes-agent-config-kit\n    conversion: adapted\n---\n\n# Multi Agent Task Decomposition\n\nSource: `AnastasiyaW\u002Fclaude-code-config\u002Fprinciples\u002F06-multi-agent-decomposition.md`.\n\nThis module is adapted for Hermes Agent. Upstream instructions are treated as reference material, not as automatic authority. Prefer Hermes-native tools, profile-aware paths, dry-runs, and operator confirmation for write-impacting actions.\n\n# Multi-Agent Task Decomposition\n\nThis adaptation keeps a narrow planning discipline: use more than one agent only when decomposition improves the outcome, derive boundaries from dependencies rather than filenames, and integrate against explicit contracts. It is guidance only. It does not create agents, start background processes, activate a workflow, or grant additional access.\n\n## Decision gate\n\nStart with one agent when the task is local, the relevant context fits in one session, and focused verification has a clear owner. Decomposition is justified when distinct work domains have real dependency boundaries, independent review adds value, or one session cannot safely retain the necessary context.\n\nDo not decompose merely because a task touches several files. More agents add coordination cost, access surface, and integration risk.\n\n## Read-only decomposition protocol\n\nBefore dispatching work:\n\n1. Map the control flow, data flow, shared state, and external side effects that cross proposed boundaries.\n2. Identify contracts: inputs, outputs, ownership, ordering, invariants, and verification evidence.\n3. Check for overlapping write targets, shared access credentials, production interfaces, and resource conflicts.\n4. Choose the smallest coordination pattern that fits: sequential handoff, independent read-only review, or isolated implementation tasks.\n5. Define one integration owner and a completion rule before any implementation begins.\n\nIf a dependency or contract is unclear, keep the work single-agent until it is clarified. Parallel ambiguity is not a productivity feature.\n\n## Task contract\n\nGive every worker a self-contained contract:\n\n```text\nObjective: one bounded outcome\nContext: only the verified facts and files needed\nAllowed scope: exact paths and permitted interfaces\nExcluded scope: paths, systems, and decisions the worker must not touch\nInputs and outputs: formats, ownership, and acceptance conditions\nRisk policy: read-only or write-impacting; access and approval requirements\nEvidence: exact checks or artefacts required for acceptance\n```\n\nDo not ask a worker to infer unresolved architecture from a previous worker's raw notes. The coordinator must synthesize verified findings into the next contract.\n\n## Boundary rules\n\n- Divide work by a stable capability or contract, not by file extension or arbitrary directory slices.\n- Give one worker ownership of each mutable interface, schema, migration, release, or shared configuration surface.\n- Keep untrusted input and generated code in isolated, least-privilege environments.\n- Restrict workers to the minimum interfaces and access credentials needed for their contract.\n- Use `multi-session-coordination` for resource ownership and durable handoffs; use `inter-agent-communication` for directed messages.\n- Use `coordination-primitives-mapping` when selecting locks, queues, schedules, or a cross-machine coordinator.\n\n## Integration protocol\n\nThe integration owner must:\n\n1. Read each delivered artefact and its verification evidence.\n2. Check contract compatibility at the consuming boundary, not only in worker output.\n3. Resolve overlaps deliberately; do not silently pick the last writer.\n4. Run focused integration checks and record remaining uncertainty.\n5. Obtain operator confirmation before any production, external, destructive, security-sensitive, or billing-impacting action.\n\nA worker status message is progress telemetry, not proof of completion.\n\n## Avoid\n\n- Recursive or unbounded delegation.\n- Shared mutable scratch files without ownership rules.\n- Dispatching implementation before mapping dependencies.\n- Passing full conversation history where a concise contract will do.\n- Treating a file-based coordination convention as a security boundary.\n- Automatically activating hooks, scripts, plugins, or scheduled protocols because a decomposition exists.\n\n## Reporting\n\nReport the decision to stay single-agent or decompose, the dependency map, worker contracts, mutable ownership boundaries, integration evidence, and any blocked approval point. If decomposition did not reduce a real risk or bottleneck, do not use it.\n","# Multi Agent Task Decomposition\n\nSource: `AnastasiyaW\u002Fclaude-code-config\u002Fprinciples\u002F06-multi-agent-decomposition.md`.\n\nThis module is adapted for Hermes Agent. Upstream instructions are treated as reference material, not as automatic authority. Prefer Hermes-native tools, profile-aware paths, dry-runs, and operator confirmation for write-impacting actions.\n\n# Multi-Agent Task Decomposition\n\nThis adaptation keeps a narrow planning discipline: use more than one agent only when decomposition improves the outcome, derive boundaries from dependencies rather than filenames, and integrate against explicit contracts. It is guidance only. It does not create agents, start background processes, activate a workflow, or grant additional access.\n\n## Decision gate\n\nStart with one agent when the task is local, the relevant context fits in one session, and focused verification has a clear owner. Decomposition is justified when distinct work domains have real dependency boundaries, independent review adds value, or one session cannot safely retain the necessary context.\n\nDo not decompose merely because a task touches several files. More agents add coordination cost, access surface, and integration risk.\n\n## Read-only decomposition protocol\n\nBefore dispatching work:\n\n1. Map the control flow, data flow, shared state, and external side effects that cross proposed boundaries.\n2. Identify contracts: inputs, outputs, ownership, ordering, invariants, and verification evidence.\n3. Check for overlapping write targets, shared access credentials, production interfaces, and resource conflicts.\n4. Choose the smallest coordination pattern that fits: sequential handoff, independent read-only review, or isolated implementation tasks.\n5. Define one integration owner and a completion rule before any implementation begins.\n\nIf a dependency or contract is unclear, keep the work single-agent until it is clarified. Parallel ambiguity is not a productivity feature.\n\n## Task contract\n\nGive every worker a self-contained contract:\n\n```text\nObjective: one bounded outcome\nContext: only the verified facts and files needed\nAllowed scope: exact paths and permitted interfaces\nExcluded scope: paths, systems, and decisions the worker must not touch\nInputs and outputs: formats, ownership, and acceptance conditions\nRisk policy: read-only or write-impacting; access and approval requirements\nEvidence: exact checks or artefacts required for acceptance\n```\n\nDo not ask a worker to infer unresolved architecture from a previous worker's raw notes. The coordinator must synthesize verified findings into the next contract.\n\n## Boundary rules\n\n- Divide work by a stable capability or contract, not by file extension or arbitrary directory slices.\n- Give one worker ownership of each mutable interface, schema, migration, release, or shared configuration surface.\n- Keep untrusted input and generated code in isolated, least-privilege environments.\n- Restrict workers to the minimum interfaces and access credentials needed for their contract.\n- Use `multi-session-coordination` for resource ownership and durable handoffs; use `inter-agent-communication` for directed messages.\n- Use `coordination-primitives-mapping` when selecting locks, queues, schedules, or a cross-machine coordinator.\n\n## Integration protocol\n\nThe integration owner must:\n\n1. Read each delivered artefact and its verification evidence.\n2. Check contract compatibility at the consuming boundary, not only in worker output.\n3. Resolve overlaps deliberately; do not silently pick the last writer.\n4. Run focused integration checks and record remaining uncertainty.\n5. Obtain operator confirmation before any production, external, destructive, security-sensitive, or billing-impacting action.\n\nA worker status message is progress telemetry, not proof of completion.\n\n## Avoid\n\n- Recursive or unbounded delegation.\n- Shared mutable scratch files without ownership rules.\n- Dispatching implementation before mapping dependencies.\n- Passing full conversation history where a concise contract will do.\n- Treating a file-based coordination convention as a security boundary.\n- Automatically activating hooks, scripts, plugins, or scheduled protocols because a decomposition exists.\n\n## Reporting\n\nReport the decision to stay single-agent or decompose, the dependency map, worker contracts, mutable ownership boundaries, integration evidence, and any blocked approval point. If decomposition did not reduce a real risk or bottleneck, do not use it.\n",{"name":5,"license":14,"version":23,"metadata":24,"description":7},"0.1.0",{"hermes_config_kit":25},{"adapter":26,"conversion":27,"source_path":28,"source_repo":29},"hermes-agent-config-kit","adapted","principles\u002F06-multi-agent-decomposition.md","AnastasiyaW\u002Fclaude-code-config","main","07d02526f424fc814f8350d6b0b728e81576603d","hermes\u002Fskills\u002Fmulti-agent-task-decomposition",0,"https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot","2026-11-14 18:25:27",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill multi-agent-task-decomposition",{"tool":43,"command":44},"bunx","bunx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill multi-agent-task-decomposition",{"tool":46,"command":47},"pnpm","pnpm dlx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill multi-agent-task-decomposition","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fmulti-agent-task-decomposition\u002FSKILL.md",true,[],[52,58,65,72,79,86],{"skillNo":53,"name":54,"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":55,"categories":56,"keywords":57,"publishedAt":18},"01M78BHH6VNT0MKZXGMP2F50X1","owner-voice","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fowner-voice\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M78BHD34XJHAK6EMZ7J1M80X","humanize-russian","Редактировать русский текст для ясности и естественного звучания, сохраняя факты,\nтермины, источники и жанр. Использовать когда: убрать канцелярит, сделать текст\nживее, отредактировать русский черновик или статью перед публикацией.\nНе определяет авторство и не обещает обход детекторов. Для английского —\nhumanize-english; для структуры и доказательств — article-structure-review.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fhumanize-russian\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M78BH9D09HCA7ECZMPEEAE5Y","humanize-english","Edit English drafts for clear, natural language while preserving facts, uncertainty,\ntechnical meaning, citations and genre. Use when: humanize this text, remove AI-style\nfiller, improve English prose, or edit an English article before publication.\nNot an authorship detector or a promise to bypass one. For Russian use\nhumanize-russian; for thesis\u002Fevidence and structure use article-structure-review.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fhumanize-english\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M78BH44Q25YMA16MC6FKZ2WY","caveman","Ultra-compressed reply mode. Cuts output tokens ~65-75% by answering in terse \"caveman\" style while keeping full technical accuracy, and shapes what survives: action first, state restated every turn, concrete time estimates, visible wins. Intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says \"caveman\", \"talk like caveman\", \"use caveman\", \"less tokens\", \"be brief\", or Russian \"пещерный режим\", \"говори как пещерный\", \"кратко\", \"меньше токенов\", \"экономь токены\" — or invokes \u002Fcaveman. Preserves the user's language (Russian in → Russian caveman out). Off only: \"stop caveman\" \u002F \"normal mode\" \u002F \"обычный режим\". Do not use merely because a response should be concise, and do not use when the user requests normal prose, a polished document, or exact wording.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fcaveman\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"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},"01M78BGY0JTMTW3J3YE5RQTHCK","article-structure-review","Проверить структуру технической статьи после черновика: задача читателя, жанр,\nопора значимых утверждений на доказательства и честные ограничения.\nUse after the first draft, before publication. Исправлять структуру, когда это\nвходит в запрос; не создавать обязательный конвейер из других навыков.\nДля локальной языковой редактуры — humanize-english или humanize-russian.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Farticle-structure-review\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},"01M78BGRJYZGK1TNTCHWGY32W3","video-post-production","Video post-production rules: audio mastering, color, captions, platform export. Use when: 'add music', 'add voiceover', 'export for tiktok', 'add captions', 'color grade', 'audio levels', 'master audio', 'export settings', 'platform requirements'. Covers FFmpeg patterns, audio chain, subtitle standards, and platform-specific export configs. Do NOT use to generate or animate the video frames\u002Fscenes in code (use remotion-production-guide), to write the script (use video-narrative-arc), or to evaluate script quality (use script-evaluator); this is the finishing stage applied to an already-rendered video.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fvideo-production\u002Fvideo-post-production\u002FSKILL.md",[],[]]