[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78AZR1K3F65JC90BTHHCP6Z":3,"skill-same:01M78AZR1K3F65JC90BTHHCP6Z":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},"01M78AZR1K3F65JC90BTHHCP6Z","feature-layer-architecture",null,"Organize long-running project knowledge into layers and feature narratives that preserve rationale, evidence, and history without replacing machine state.","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\u002Ffeature-layer-architecture\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: feature-layer-architecture\ndescription: \"Organize long-running project knowledge into layers and feature narratives that preserve rationale, evidence, and history without replacing machine state.\"\nversion: 0.1.0\nlicense: MIT\nmetadata:\n  hermes_config_kit:\n    source_repo: AnastasiyaW\u002Fclaude-code-config\n    source_path: principles\u002F28-feature-layer-architecture.md\n    adapter: hermes-agent-config-kit\n    conversion: adapted\n---\n\n# Feature Layer Architecture\n\nSource: `AnastasiyaW\u002Fclaude-code-config\u002Fprinciples\u002F28-feature-layer-architecture.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# Feature Layer Architecture\n\nUpstream source policy describes a three-tier knowledge model for long-running projects. Hermes adaptation keeps the architectural pattern — global principles, project layers, and feature narratives.\n\nConcrete scaffolding tooling for this pattern is also ported: the `layer-new` and `feature-new` skills mechanically create the directory tree and feature narratives described below, drawing from the installed `kb-skeleton` template; `build_kb_graph.py` and `validate_kb.py` (reviewed scripts bundled with that template) check consistency once a project adopts it. This module stays the conceptual\u002Fadoption-judgment layer — use `layer-new`\u002F`feature-new` for the mechanical part, and see this module for whether and how to structure it.\n\n## Principle\n\nOrganize long-running project knowledge into layers and feature narratives when machine state alone no longer preserves design rationale.\n\nUse this module when a project has enough history that `feature_list.json`, handoffs, and commit logs tell what happened, but not why the current shape exists.\n\n## Three-tier model\n\nUse a three-tier tree:\n\n1. **Global knowledge** — reusable principles, rules, and modules that transfer across projects.\n2. **Project layer knowledge** — bounded concerns inside one project: security, data, infrastructure, UI, domain logic, operations, or integration boundaries.\n3. **Feature narratives** — per-feature design, plan, verification evidence, deviations, and conclusion.\n\nThe tiers have different jobs. Do not collapse them into one mega-document.\n\n## What is a layer?\n\nA layer is a bounded concern, not merely a folder name.\n\nExamples:\n\n- security and access control;\n- data model and persistence;\n- user interface and interaction design;\n- infrastructure and deployment;\n- external integrations;\n- domain logic;\n- operational runbooks.\n\nA file may participate in multiple layers. A feature has one primary layer and may explicitly touch secondary layers.\n\n## Recommended structure\n\nFor projects that earn the overhead, keep layer material under a predictable project-local location such as:\n\n```text\ndocs\u002Flayers\u002F\u003Clayer-name>\u002F\n  README.md\n  kb\u002F\n    invariants.md\n    decisions.md\n    gotchas.md\n    patterns.md\n  history.md\n  features\u002F\n    feat-NNN-\u003Cslug>.md\n```\n\nThis is a convention, not a command to create directories blindly. Start with the smallest layer tree that helps future work.\n\n## Layer README\n\nEach layer entry point should state:\n\n- purpose;\n- status: active, deprecated, merging, or archived;\n- governing principles and project rules;\n- local invariants summary;\n- feature index;\n- dependencies on other layers;\n- where verification evidence lives.\n\n## Layer knowledge base\n\nLayer-local KB files should separate different kinds of knowledge:\n\n- **invariants** — rules that must remain true for this layer;\n- **decisions** — architectural decisions and rejected alternatives;\n- **gotchas** — pitfalls, incident lessons, and sharp edges;\n- **patterns** — reusable recipes that have survived verification.\n\nIf a layer-local pattern is reused across projects, promote it deliberately into a global principle or module. Promotion should be earned by usage, not optimism.\n\n## Feature narrative\n\nA feature narrative should preserve:\n\n- feature ID and title;\n- primary layer and touched layers;\n- status;\n- related feature IDs;\n- design rationale;\n- assumptions and unknowns;\n- plan and phases;\n- files and interfaces touched;\n- verification evidence;\n- deviations from plan;\n- conclusion and future work.\n\nWhen the feature is done, close the narrative as history. Do not keep rewriting old feature documents to pretend the original plan was perfect. New work gets a new feature or superseding note.\n\n## Relationship to machine state\n\nUse `long-run-feature-tracking` for machine-readable state: IDs, status, dependencies, and evidence pointers.\n\nUse feature-layer architecture for human-readable rationale: why this layer exists, why a feature took its shape, what alternatives were rejected, and what should not be rediscovered six weeks later.\n\nThe two should cite each other, but not duplicate each other.\n\n## Adoption threshold\n\nThis earns its complexity when the project has:\n\n- multiple months of work;\n- five or more active concerns;\n- multiple sessions or collaborators;\n- recurring confusion about why code is shaped a certain way;\n- cross-cutting features that touch more than one concern;\n- verified decisions that keep getting rediscovered.\n\nSkip it for:\n\n- short-lived utilities;\n- one-off migrations;\n- prototypes or spikes;\n- projects with only a few features;\n- teams that will not maintain the documents.\n\nDocumentation nobody updates is not architecture. It is sediment.\n\n## Adoption protocol\n\n1. Identify the few bounded concerns that currently cause navigation pain.\n2. Create only those layer entries.\n3. For each layer, write the README first: purpose, invariants, active features, dependencies.\n4. Move or link existing durable evidence rather than rewriting history from memory.\n5. Add feature narratives only for active or high-value completed features.\n6. Cross-link to `feature_list.json`, issue trackers, commits, and verification artefacts.\n7. Add validation only after the manual convention is stable.\n\n## Review checklist\n\nBefore adopting or expanding this structure, verify:\n\n- [ ] The project is long-running enough to justify the overhead.\n- [ ] Each layer is a bounded concern, not a renamed directory.\n- [ ] Machine state and human narrative are not duplicated.\n- [ ] Feature documents have clear ownership and closure rules.\n- [ ] Layer history is append-only or otherwise auditable.\n- [ ] Links point to durable artefacts rather than transient chat.\n- [ ] Promotion from feature to layer to global knowledge is based on reuse.\n\n## Avoid\n\n- Creating a full layer tree before there are real layers.\n- Writing layer documentation as a substitute for tests, issues, or feature state.\n- Baking project-local paths into global rules.\n- Letting feature docs become mutable status dashboards.\n- Treating old chat transcripts as durable rationale.\n- Adding validators before the information model is stable.\n\n## Reporting format\n\nWhen using this module, report:\n\n- project maturity signal;\n- proposed layers;\n- feature narratives to create or migrate;\n- what remains in machine-readable state;\n- what becomes layer knowledge;\n- validation plan, if any;\n- overhead intentionally avoided.\n\nThe goal is not more documents. The goal is to make the project’s memory navigable without asking the same questions every month.\n","# Feature Layer Architecture\n\nSource: `AnastasiyaW\u002Fclaude-code-config\u002Fprinciples\u002F28-feature-layer-architecture.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# Feature Layer Architecture\n\nUpstream source policy describes a three-tier knowledge model for long-running projects. Hermes adaptation keeps the architectural pattern — global principles, project layers, and feature narratives.\n\nConcrete scaffolding tooling for this pattern is also ported: the `layer-new` and `feature-new` skills mechanically create the directory tree and feature narratives described below, drawing from the installed `kb-skeleton` template; `build_kb_graph.py` and `validate_kb.py` (reviewed scripts bundled with that template) check consistency once a project adopts it. This module stays the conceptual\u002Fadoption-judgment layer — use `layer-new`\u002F`feature-new` for the mechanical part, and see this module for whether and how to structure it.\n\n## Principle\n\nOrganize long-running project knowledge into layers and feature narratives when machine state alone no longer preserves design rationale.\n\nUse this module when a project has enough history that `feature_list.json`, handoffs, and commit logs tell what happened, but not why the current shape exists.\n\n## Three-tier model\n\nUse a three-tier tree:\n\n1. **Global knowledge** — reusable principles, rules, and modules that transfer across projects.\n2. **Project layer knowledge** — bounded concerns inside one project: security, data, infrastructure, UI, domain logic, operations, or integration boundaries.\n3. **Feature narratives** — per-feature design, plan, verification evidence, deviations, and conclusion.\n\nThe tiers have different jobs. Do not collapse them into one mega-document.\n\n## What is a layer?\n\nA layer is a bounded concern, not merely a folder name.\n\nExamples:\n\n- security and access control;\n- data model and persistence;\n- user interface and interaction design;\n- infrastructure and deployment;\n- external integrations;\n- domain logic;\n- operational runbooks.\n\nA file may participate in multiple layers. A feature has one primary layer and may explicitly touch secondary layers.\n\n## Recommended structure\n\nFor projects that earn the overhead, keep layer material under a predictable project-local location such as:\n\n```text\ndocs\u002Flayers\u002F\u003Clayer-name>\u002F\n  README.md\n  kb\u002F\n    invariants.md\n    decisions.md\n    gotchas.md\n    patterns.md\n  history.md\n  features\u002F\n    feat-NNN-\u003Cslug>.md\n```\n\nThis is a convention, not a command to create directories blindly. Start with the smallest layer tree that helps future work.\n\n## Layer README\n\nEach layer entry point should state:\n\n- purpose;\n- status: active, deprecated, merging, or archived;\n- governing principles and project rules;\n- local invariants summary;\n- feature index;\n- dependencies on other layers;\n- where verification evidence lives.\n\n## Layer knowledge base\n\nLayer-local KB files should separate different kinds of knowledge:\n\n- **invariants** — rules that must remain true for this layer;\n- **decisions** — architectural decisions and rejected alternatives;\n- **gotchas** — pitfalls, incident lessons, and sharp edges;\n- **patterns** — reusable recipes that have survived verification.\n\nIf a layer-local pattern is reused across projects, promote it deliberately into a global principle or module. Promotion should be earned by usage, not optimism.\n\n## Feature narrative\n\nA feature narrative should preserve:\n\n- feature ID and title;\n- primary layer and touched layers;\n- status;\n- related feature IDs;\n- design rationale;\n- assumptions and unknowns;\n- plan and phases;\n- files and interfaces touched;\n- verification evidence;\n- deviations from plan;\n- conclusion and future work.\n\nWhen the feature is done, close the narrative as history. Do not keep rewriting old feature documents to pretend the original plan was perfect. New work gets a new feature or superseding note.\n\n## Relationship to machine state\n\nUse `long-run-feature-tracking` for machine-readable state: IDs, status, dependencies, and evidence pointers.\n\nUse feature-layer architecture for human-readable rationale: why this layer exists, why a feature took its shape, what alternatives were rejected, and what should not be rediscovered six weeks later.\n\nThe two should cite each other, but not duplicate each other.\n\n## Adoption threshold\n\nThis earns its complexity when the project has:\n\n- multiple months of work;\n- five or more active concerns;\n- multiple sessions or collaborators;\n- recurring confusion about why code is shaped a certain way;\n- cross-cutting features that touch more than one concern;\n- verified decisions that keep getting rediscovered.\n\nSkip it for:\n\n- short-lived utilities;\n- one-off migrations;\n- prototypes or spikes;\n- projects with only a few features;\n- teams that will not maintain the documents.\n\nDocumentation nobody updates is not architecture. It is sediment.\n\n## Adoption protocol\n\n1. Identify the few bounded concerns that currently cause navigation pain.\n2. Create only those layer entries.\n3. For each layer, write the README first: purpose, invariants, active features, dependencies.\n4. Move or link existing durable evidence rather than rewriting history from memory.\n5. Add feature narratives only for active or high-value completed features.\n6. Cross-link to `feature_list.json`, issue trackers, commits, and verification artefacts.\n7. Add validation only after the manual convention is stable.\n\n## Review checklist\n\nBefore adopting or expanding this structure, verify:\n\n- [ ] The project is long-running enough to justify the overhead.\n- [ ] Each layer is a bounded concern, not a renamed directory.\n- [ ] Machine state and human narrative are not duplicated.\n- [ ] Feature documents have clear ownership and closure rules.\n- [ ] Layer history is append-only or otherwise auditable.\n- [ ] Links point to durable artefacts rather than transient chat.\n- [ ] Promotion from feature to layer to global knowledge is based on reuse.\n\n## Avoid\n\n- Creating a full layer tree before there are real layers.\n- Writing layer documentation as a substitute for tests, issues, or feature state.\n- Baking project-local paths into global rules.\n- Letting feature docs become mutable status dashboards.\n- Treating old chat transcripts as durable rationale.\n- Adding validators before the information model is stable.\n\n## Reporting format\n\nWhen using this module, report:\n\n- project maturity signal;\n- proposed layers;\n- feature narratives to create or migrate;\n- what remains in machine-readable state;\n- what becomes layer knowledge;\n- validation plan, if any;\n- overhead intentionally avoided.\n\nThe goal is not more documents. The goal is to make the project’s memory navigable without asking the same questions every month.\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\u002F28-feature-layer-architecture.md","AnastasiyaW\u002Fclaude-code-config","main","07d02526f424fc814f8350d6b0b728e81576603d","hermes\u002Fskills\u002Ffeature-layer-architecture",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 feature-layer-architecture",{"tool":43,"command":44},"bunx","bunx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill feature-layer-architecture",{"tool":46,"command":47},"pnpm","pnpm dlx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill feature-layer-architecture","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Ffeature-layer-architecture\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",[],[]]