[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78B91ESTBDC76H66VDJ4R5N":3,"skill-same:01M78B91ESTBDC76H66VDJ4R5N":52},{"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":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":32,"install":33,"resources":45},"01M78B91ESTBDC76H66VDJ4R5N","feature-new",null,"Scaffold a feature narrative in an existing project layer with Design, Plan, Verify, and Conclusion sections. Use when: \"create a new feature\", \"start work on feature\", \"scaffold feature doc\", \"\u002Ffeature-new\", or \"begin feature narrative\". Inspects and preserves the target project's feature ID, registry, filename, and coordination conventions; does not create a missing layer (use \u002Flayer-new).","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\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Farchitecture\u002Ffeature-new\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: feature-new\ndescription: >\n  Scaffold a feature narrative in an existing project layer with Design,\n  Plan, Verify, and Conclusion sections. Use when: \"create a new feature\",\n  \"start work on feature\", \"scaffold feature doc\", \"\u002Ffeature-new\", or\n  \"begin feature narrative\". Inspects and preserves the target project's\n  feature ID, registry, filename, and coordination conventions; does not\n  create a missing layer (use \u002Flayer-new).\nuser-invocable: true\nmodel: sonnet\n---\n\n# \u002Ffeature-new -- scaffold a feature narrative\n\nCreates a feature document in an existing layer. The document follows\nthe ULTRAPACK-style narrative template (Design \u002F Plan \u002F Verify \u002F\nConclusion) extended with explicit cross-references to layer\ninvariants and global principles.\n\n## When to use\n\n- Beginning design work on a new feature, **before** writing code\n- Migrating an in-flight feature from \"scattered context\" into the\n  formal narrative\n- Creating a feature placeholder when planning future work that\n  another session will pick up\n\n## When NOT to use\n\n- One-line bug fixes that do not need a design phase (just commit)\n- Documentation-only changes (those go in handoffs or PR\n  descriptions)\n- Refactors with no behavioral change (commit message is sufficient)\n\n## Arguments\n\n```\n\u002Ffeature-new \u003Clayer> \u003Cslug> [--title \"...\"] [--branch \u003Cname>] [--id \u003Cproject-id>]\n```\n\n- `\u003Clayer>` -- existing layer name. Must be a directory under\n  `docs\u002Flayers\u002F`. If missing, suggest `\u002Flayer-new \u003Clayer>` first.\n- `\u003Cslug>` -- kebab-case feature identifier without a project ID\n  prefix. Examples: `api-key-rotation`, `audit-log`,\n  `dual-encryption`.\n- `--title` -- human-readable feature title. If omitted, derive from\n  slug by title-casing.\n- `--branch` -- git branch name. If omitted, default to\n  `feature\u002F\u003Cslug>`.\n- `--id` -- override the auto-allocated project-native ID. Use only when\n  migrating a pre-existing feature with a known ID. Refuse if the ID\n  already exists in this layer.\n\n## Direction (what to do, in order)\n\n### Step 1 -- Verify environment\n\n1. Determine repo root via `git rev-parse --show-toplevel`.\n2. Confirm `docs\u002Flayers\u002F\u003Clayer>\u002F` exists. If not, refuse with a\n   suggestion to run `\u002Flayer-new \u003Clayer>` first.\n3. Confirm `docs\u002Flayers\u002F\u003Clayer>\u002Ffeatures\u002F_FEATURE-TEMPLATE.md`\n   exists. If not, copy from\n   `${CLAUDE_PLUGIN_ROOT}\u002Ftemplates\u002Fkb-skeleton\u002Fdocs\u002Flayers\u002F_LAYER-TEMPLATE\u002Ffeatures\u002F_FEATURE-TEMPLATE.md`.\n\n### Step 2 -- Discover and reserve the project-native ID\n\nDo this before changing the feature document, layer README, or registry. A\nfeature registry is project-owned state; this skill does not impose its old\n`F-NNN` example on it.\n\n1. Inspect existing feature documents, the layer README, `feature_list.json`\n   (if present), a co-located schema or validator, and `AGENTS.md`\u002Fproject\n   docs for the actual ID format, namespace, filename convention, fields,\n   initial status, allocator, and coordination rule. For example, this\n   repository's long-run template uses lowercase `feat-NNN` IDs with\n   `description`, `dependencies`, and string `evidence`; that is not\n   interchangeable with the old `F-NNN` example.\n2. If `--id` was provided, validate it against the discovered convention and\n   check every project-owned feature source that the convention names. If it\n   is already used, refuse without changing anything.\n3. Otherwise use the project's allocator when one exists. If the convention is\n   visible only in current records, derive a candidate from those records and\n   recheck all of them immediately before reservation. Do not infer a global\n   namespace, digit width, prefix, or sort order from this skill.\n4. Follow an existing project coordination mechanism when one exists and you\n   are authorized to use it. Otherwise reserve the **logical ID**, not a\n   slug-bearing pathname, with `scripts\u002Freserve_feature_id.py`. It atomically\n   creates `refs\u002Ffeature-new\u002Freservations\u002F\u003Cproject-id>` in the repository and\n   binds that ID to the layer, slug, and final document path. This is a Git\n   allocator record, not a second feature registry: the same request returns\n   `resumed`; a different layer\u002Fslug\u002Fpath returns `conflict` and must not edit\n   the registry or another claimant's document.\n5. On `conflict`, re-inventory the project convention and choose the next\n   project-native candidate. Continue safe reconciliation while the requested\n   scaffold remains actionable; do not guess that a durable reservation has a\n   stale owner, overwrite it, or impose an arbitrary retry limit.\n6. If no project ID\u002Fpath convention exists at all, bootstrap only this skill's\n   documented default: lowercase `feat-NNN`, globally allocated through the\n   Git-ref helper, with `docs\u002Flayers\u002F\u003Clayer>\u002Ffeatures\u002Ffeat-NNN-\u003Cslug>.md` as\n   the document path. Do not create `feature_list.json`; once a registry exists\n   it becomes authoritative over this fallback.\n7. After reservation, re-read the README and registry before companion\n   updates. If their convention changed, preserve the valid document and stop\n   before a stale README\u002Fregistry write; report the exact conflict and reserved\n   path for the owner to reconcile.\n\nWhen the project has no registry schema or helper, do not stop merely because\nit differs from this skill's example. Adapt by copying one current entry's\nfield set and ID style, changing only values whose meaning is established by\nthe project. If no representative entry or documented meaning exists, create\nthe requested narrative only and leave the unknown registry untouched rather\nthan append an invented record.\n\n### Step 3 -- Validate slug\n\n- Lowercase kebab-case (`[a-z][a-z0-9-]*`).\n- Length \u003C= 50 characters.\n- Does not repeat the discovered ID prefix.\n- The resulting project-native feature filename does not already exist.\n\n### Step 4 -- Copy and fill the template\n\nSource: `docs\u002Flayers\u002F\u003Clayer>\u002Ffeatures\u002F_FEATURE-TEMPLATE.md`\n\nDestination: the project-native feature filename discovered in Step 2.\n\nIn the new file, replace placeholders:\n\n| Placeholder | Replacement |\n|-------------|-------------|\n| feature ID\u002Ftitle placeholder | the discovered ID and `\u003Ctitle>` |\n| `**Layer:** [\u003Clayer-name>](..\u002FREADME.md)` | `**Layer:** [\u003Clayer>](..\u002FREADME.md)` |\n| `**Status:** design` | leave as `design` |\n| `**Branch:** feature\u002F\u003Cslug>` | use `--branch` value or default |\n| `**Started:** YYYY-MM-DD` | today's date |\n| `**Owner:** \u003Cname>` | infer from git config user.name, or leave placeholder |\n\nLeave Design \u002F Plan \u002F Verify \u002F Conclusion section bodies as template\nplaceholders -- the user fills these.\n\n### Step 5 -- Update layer README\n\nInspect `docs\u002Flayers\u002F\u003Clayer>\u002FREADME.md` for its existing feature index and\npreserve its columns, ID form, ordering, and link style. Insert the new entry\nonly if a feature index exists and its row convention is understood:\n\n```\n| \u003Cproject-id> | \u003Ctitle> | \u003Cproject-native initial status> | YYYY-MM-DD | \u003Cproject-native link> |\n```\n\nIf the table has only known placeholder rows, replace only those placeholders.\nIf there is no compatible index, do not invent one; report that the narrative\nwas created without a README index.\n\n### Step 6 -- Update feature_list.json (if present)\n\nIf `\u003Crepo>\u002Ffeature_list.json` exists at repo root, parse it and use the\nproject's schema or validator plus a representative entry to construct a\nnative record. Preserve the existing top-level shape, field names, value\ntypes, default status, evidence representation, ordering, and encoding.\nChange only fields that the project convention establishes for a new feature\n(such as its unique ID, title\u002Fname, description, document link, or initial\nstate).\n\nImportant encoding rule (per\n`~\u002F.claude\u002Frules\u002Fapi-utf8-posting.md`): write the JSON file with\n`json.dump(data, f, ensure_ascii=False, indent=2)` to preserve any\nCyrillic in titles.\n\nDo NOT change existing entries or add fields merely because this skill's old\nexample had them. Run the project-provided feature registry validator when one\nexists. If the registry's required fields cannot be determined safely, leave\nit unchanged and report that fact; the requested narrative remains valid.\n\nIf `feature_list.json` does not exist, do not auto-create it -- emit\na hint instead.\n\n### Step 7 -- Confirm and suggest next step\n\nPrint a summary:\n\n```\nCreated: \u003Cproject-native feature-document path>\nUpdated: \u003CREADME path or \"not indexed; no compatible feature index\">\nUpdated: \u003Cfeature_list.json path and project-native ID, or \"not changed; no safe registry mapping\">\n\nSuggested next steps:\n1. Fill the Design section in \u003Cfeature document>\n   - Approach (one paragraph)\n   - Invariants (IV-1, IV-2, ...)\n   - Rejected alternatives\n2. When Design is reviewed, change Status: design -> planning and fill Plan\n3. Create the git branch: git checkout -b feature\u002F\u003Cslug>\n```\n\n## Blueprints (files this skill writes from)\n\n- `${CLAUDE_PLUGIN_ROOT}\u002Ftemplates\u002Fkb-skeleton\u002Fdocs\u002Flayers\u002F_LAYER-TEMPLATE\u002Ffeatures\u002F_FEATURE-TEMPLATE.md` -- the source template\n\n## Status lifecycle\n\nThe following lifecycle is the bundled long-run-project convention, not a\nuniversal registry schema. Apply it only after Step 2 confirms that the target\nproject uses it; otherwise preserve the project's own lifecycle and mapping.\n\n### Doc Status (narrative phase, in feature.md frontmatter)\n\nTracks where in the ULTRAPACK Design \u002F Plan \u002F Verify \u002F Conclusion\njourney the feature is.\n\n```\ndesign --> planning --> executing --> reviewing --> done\n                                  \\\n                                   --> blocked --> executing\n```\n\nSix states: `design`, `planning`, `executing`, `reviewing`, `done`,\n`blocked`. Transitions are manual edits. Once `done`, the feature doc\nis read-only history; further changes go into a superseding feature.\n\n### feature_list.json status (machine state, for tooling)\n\nTracks the machine-readable state used by build_kb_graph.py and\nvalidate_kb_links.py.\n\n```\nnot-started --> in-progress --> done\n              \\\n               --> blocked --> in-progress\n```\n\nFour states: `not-started`, `in-progress`, `blocked`, `done`. `done` is\n**one-way** (no rollback; regression becomes a new feature) per\n[principle 27](https:\u002F\u002Fgithub.com\u002FAnastasiyaW\u002Fcodex-claude-code-config\u002Fblob\u002Fmain\u002Fprinciples\u002F27-feature-tracking.md).\n\n### Mapping between the two\n\n| Doc Status | feature_list.json status | Notes |\n|------------|--------------------------|-------|\n| design | not-started | newly created, no plan yet |\n| planning | in-progress | plan being written |\n| executing | in-progress | code being written |\n| reviewing | in-progress | review\u002Fverify phase |\n| blocked | blocked | identical |\n| done | done | identical |\n\nFor a project that uses this convention, create the doc with `Status: design`\nand the native JSON record with `status: \"not-started\"`. Subsequent transitions\nare manual and follow that project's coordination rule.\n\n## Gotchas\n\n- **ID namespace is project-defined.** It may be global, per layer, or managed\n  by a registry. Discover it before allocation; never treat `F-NNN` as a\n  universal format.\n- **Concurrent allocation.** Scanning and then writing is a race. Use the\n  existing project coordinator or reserve the logical ID with the Git-ref\n  helper before a document\u002Fregistry mutation. A same-ID different-slug\u002Flayer\n  claimant must conflict; re-inventory it, never overwrite it or guess that\n  the durable reservation is stale.\n- **Migration of in-flight features.** When migrating an existing\n  feature into the new format, pass its project-native `--id` explicitly so the\n  feature retains its prior ID in any links from PROBLEMS.md or\n  handoffs. The skill will not auto-detect existing IDs.\n- **Cyrillic titles + Windows.** Per global rule\n  `api-utf8-posting.md`, when writing the markdown file or\n  feature_list.json, always specify `encoding=\"utf-8\"` explicitly to\n  avoid mojibake on Windows.\n- **Layer README table edit.** Preserve the project's existing columns and\n  ordering. If the index cannot be interpreted safely, leave it unchanged and\n  report the unindexed narrative rather than inventing a canonical 5-column\n  row.\n\n## Troubleshooting\n\n| Symptom | Cause | Fix |\n|---------|-------|-----|\n| \"Layer does not exist\" | `docs\u002Flayers\u002F\u003Clayer>\u002F` missing | Run `\u002Flayer-new \u003Clayer>` first |\n| ID reservation conflict | Another request owns that logical ID | Re-inventory project state and use the next native candidate; never overwrite or delete the durable reservation |\n| `feature_list.json` parse\u002Fschema error | Registry is invalid or uses an unknown convention | Preserve the narrative and do not write the registry; report the exact parser\u002Fschema failure and use a project-native adapter when its fields are established |\n| Template missing on this machine | Different host \u002F fresh clone | Pull from public repo: `gh api repos\u002FAnastasiyaW\u002Fcodex-claude-code-config\u002Fcontents\u002Ftemplates\u002Fkb-skeleton\u002Fdocs\u002Flayers\u002F_LAYER-TEMPLATE\u002Ffeatures\u002F_FEATURE-TEMPLATE.md` |\n| Cyrillic in title shows as `?????` | File written without explicit utf-8 | Re-write the file with `encoding=\"utf-8\"`; see `~\u002F.claude\u002Frules\u002Fapi-utf8-posting.md` |\n\n## Implementation note\n\nThis is a **scaffolding** skill: file copy + placeholder replacement +\nsmall JSON merge. Keep it deterministic. The Design \u002F Plan \u002F Verify\nsections of the produced document are meant for the user (or the\nsession that invoked the skill) to fill -- this skill does not\nattempt to generate Design content from the title.\n\nID discovery reads only the sources that the project's convention declares.\nReserve the resulting logical ID atomically, then re-read mutable sources;\ncache is not authority across a concurrent mutation.\n","# \u002Ffeature-new -- scaffold a feature narrative\n\nCreates a feature document in an existing layer. The document follows\nthe ULTRAPACK-style narrative template (Design \u002F Plan \u002F Verify \u002F\nConclusion) extended with explicit cross-references to layer\ninvariants and global principles.\n\n## When to use\n\n- Beginning design work on a new feature, **before** writing code\n- Migrating an in-flight feature from \"scattered context\" into the\n  formal narrative\n- Creating a feature placeholder when planning future work that\n  another session will pick up\n\n## When NOT to use\n\n- One-line bug fixes that do not need a design phase (just commit)\n- Documentation-only changes (those go in handoffs or PR\n  descriptions)\n- Refactors with no behavioral change (commit message is sufficient)\n\n## Arguments\n\n```\n\u002Ffeature-new \u003Clayer> \u003Cslug> [--title \"...\"] [--branch \u003Cname>] [--id \u003Cproject-id>]\n```\n\n- `\u003Clayer>` -- existing layer name. Must be a directory under\n  `docs\u002Flayers\u002F`. If missing, suggest `\u002Flayer-new \u003Clayer>` first.\n- `\u003Cslug>` -- kebab-case feature identifier without a project ID\n  prefix. Examples: `api-key-rotation`, `audit-log`,\n  `dual-encryption`.\n- `--title` -- human-readable feature title. If omitted, derive from\n  slug by title-casing.\n- `--branch` -- git branch name. If omitted, default to\n  `feature\u002F\u003Cslug>`.\n- `--id` -- override the auto-allocated project-native ID. Use only when\n  migrating a pre-existing feature with a known ID. Refuse if the ID\n  already exists in this layer.\n\n## Direction (what to do, in order)\n\n### Step 1 -- Verify environment\n\n1. Determine repo root via `git rev-parse --show-toplevel`.\n2. Confirm `docs\u002Flayers\u002F\u003Clayer>\u002F` exists. If not, refuse with a\n   suggestion to run `\u002Flayer-new \u003Clayer>` first.\n3. Confirm `docs\u002Flayers\u002F\u003Clayer>\u002Ffeatures\u002F_FEATURE-TEMPLATE.md`\n   exists. If not, copy from\n   `${CLAUDE_PLUGIN_ROOT}\u002Ftemplates\u002Fkb-skeleton\u002Fdocs\u002Flayers\u002F_LAYER-TEMPLATE\u002Ffeatures\u002F_FEATURE-TEMPLATE.md`.\n\n### Step 2 -- Discover and reserve the project-native ID\n\nDo this before changing the feature document, layer README, or registry. A\nfeature registry is project-owned state; this skill does not impose its old\n`F-NNN` example on it.\n\n1. Inspect existing feature documents, the layer README, `feature_list.json`\n   (if present), a co-located schema or validator, and `AGENTS.md`\u002Fproject\n   docs for the actual ID format, namespace, filename convention, fields,\n   initial status, allocator, and coordination rule. For example, this\n   repository's long-run template uses lowercase `feat-NNN` IDs with\n   `description`, `dependencies`, and string `evidence`; that is not\n   interchangeable with the old `F-NNN` example.\n2. If `--id` was provided, validate it against the discovered convention and\n   check every project-owned feature source that the convention names. If it\n   is already used, refuse without changing anything.\n3. Otherwise use the project's allocator when one exists. If the convention is\n   visible only in current records, derive a candidate from those records and\n   recheck all of them immediately before reservation. Do not infer a global\n   namespace, digit width, prefix, or sort order from this skill.\n4. Follow an existing project coordination mechanism when one exists and you\n   are authorized to use it. Otherwise reserve the **logical ID**, not a\n   slug-bearing pathname, with `scripts\u002Freserve_feature_id.py`. It atomically\n   creates `refs\u002Ffeature-new\u002Freservations\u002F\u003Cproject-id>` in the repository and\n   binds that ID to the layer, slug, and final document path. This is a Git\n   allocator record, not a second feature registry: the same request returns\n   `resumed`; a different layer\u002Fslug\u002Fpath returns `conflict` and must not edit\n   the registry or another claimant's document.\n5. On `conflict`, re-inventory the project convention and choose the next\n   project-native candidate. Continue safe reconciliation while the requested\n   scaffold remains actionable; do not guess that a durable reservation has a\n   stale owner, overwrite it, or impose an arbitrary retry limit.\n6. If no project ID\u002Fpath convention exists at all, bootstrap only this skill's\n   documented default: lowercase `feat-NNN`, globally allocated through the\n   Git-ref helper, with `docs\u002Flayers\u002F\u003Clayer>\u002Ffeatures\u002Ffeat-NNN-\u003Cslug>.md` as\n   the document path. Do not create `feature_list.json`; once a registry exists\n   it becomes authoritative over this fallback.\n7. After reservation, re-read the README and registry before companion\n   updates. If their convention changed, preserve the valid document and stop\n   before a stale README\u002Fregistry write; report the exact conflict and reserved\n   path for the owner to reconcile.\n\nWhen the project has no registry schema or helper, do not stop merely because\nit differs from this skill's example. Adapt by copying one current entry's\nfield set and ID style, changing only values whose meaning is established by\nthe project. If no representative entry or documented meaning exists, create\nthe requested narrative only and leave the unknown registry untouched rather\nthan append an invented record.\n\n### Step 3 -- Validate slug\n\n- Lowercase kebab-case (`[a-z][a-z0-9-]*`).\n- Length \u003C= 50 characters.\n- Does not repeat the discovered ID prefix.\n- The resulting project-native feature filename does not already exist.\n\n### Step 4 -- Copy and fill the template\n\nSource: `docs\u002Flayers\u002F\u003Clayer>\u002Ffeatures\u002F_FEATURE-TEMPLATE.md`\n\nDestination: the project-native feature filename discovered in Step 2.\n\nIn the new file, replace placeholders:\n\n| Placeholder | Replacement |\n|-------------|-------------|\n| feature ID\u002Ftitle placeholder | the discovered ID and `\u003Ctitle>` |\n| `**Layer:** [\u003Clayer-name>](..\u002FREADME.md)` | `**Layer:** [\u003Clayer>](..\u002FREADME.md)` |\n| `**Status:** design` | leave as `design` |\n| `**Branch:** feature\u002F\u003Cslug>` | use `--branch` value or default |\n| `**Started:** YYYY-MM-DD` | today's date |\n| `**Owner:** \u003Cname>` | infer from git config user.name, or leave placeholder |\n\nLeave Design \u002F Plan \u002F Verify \u002F Conclusion section bodies as template\nplaceholders -- the user fills these.\n\n### Step 5 -- Update layer README\n\nInspect `docs\u002Flayers\u002F\u003Clayer>\u002FREADME.md` for its existing feature index and\npreserve its columns, ID form, ordering, and link style. Insert the new entry\nonly if a feature index exists and its row convention is understood:\n\n```\n| \u003Cproject-id> | \u003Ctitle> | \u003Cproject-native initial status> | YYYY-MM-DD | \u003Cproject-native link> |\n```\n\nIf the table has only known placeholder rows, replace only those placeholders.\nIf there is no compatible index, do not invent one; report that the narrative\nwas created without a README index.\n\n### Step 6 -- Update feature_list.json (if present)\n\nIf `\u003Crepo>\u002Ffeature_list.json` exists at repo root, parse it and use the\nproject's schema or validator plus a representative entry to construct a\nnative record. Preserve the existing top-level shape, field names, value\ntypes, default status, evidence representation, ordering, and encoding.\nChange only fields that the project convention establishes for a new feature\n(such as its unique ID, title\u002Fname, description, document link, or initial\nstate).\n\nImportant encoding rule (per\n`~\u002F.claude\u002Frules\u002Fapi-utf8-posting.md`): write the JSON file with\n`json.dump(data, f, ensure_ascii=False, indent=2)` to preserve any\nCyrillic in titles.\n\nDo NOT change existing entries or add fields merely because this skill's old\nexample had them. Run the project-provided feature registry validator when one\nexists. If the registry's required fields cannot be determined safely, leave\nit unchanged and report that fact; the requested narrative remains valid.\n\nIf `feature_list.json` does not exist, do not auto-create it -- emit\na hint instead.\n\n### Step 7 -- Confirm and suggest next step\n\nPrint a summary:\n\n```\nCreated: \u003Cproject-native feature-document path>\nUpdated: \u003CREADME path or \"not indexed; no compatible feature index\">\nUpdated: \u003Cfeature_list.json path and project-native ID, or \"not changed; no safe registry mapping\">\n\nSuggested next steps:\n1. Fill the Design section in \u003Cfeature document>\n   - Approach (one paragraph)\n   - Invariants (IV-1, IV-2, ...)\n   - Rejected alternatives\n2. When Design is reviewed, change Status: design -> planning and fill Plan\n3. Create the git branch: git checkout -b feature\u002F\u003Cslug>\n```\n\n## Blueprints (files this skill writes from)\n\n- `${CLAUDE_PLUGIN_ROOT}\u002Ftemplates\u002Fkb-skeleton\u002Fdocs\u002Flayers\u002F_LAYER-TEMPLATE\u002Ffeatures\u002F_FEATURE-TEMPLATE.md` -- the source template\n\n## Status lifecycle\n\nThe following lifecycle is the bundled long-run-project convention, not a\nuniversal registry schema. Apply it only after Step 2 confirms that the target\nproject uses it; otherwise preserve the project's own lifecycle and mapping.\n\n### Doc Status (narrative phase, in feature.md frontmatter)\n\nTracks where in the ULTRAPACK Design \u002F Plan \u002F Verify \u002F Conclusion\njourney the feature is.\n\n```\ndesign --> planning --> executing --> reviewing --> done\n                                  \\\n                                   --> blocked --> executing\n```\n\nSix states: `design`, `planning`, `executing`, `reviewing`, `done`,\n`blocked`. Transitions are manual edits. Once `done`, the feature doc\nis read-only history; further changes go into a superseding feature.\n\n### feature_list.json status (machine state, for tooling)\n\nTracks the machine-readable state used by build_kb_graph.py and\nvalidate_kb_links.py.\n\n```\nnot-started --> in-progress --> done\n              \\\n               --> blocked --> in-progress\n```\n\nFour states: `not-started`, `in-progress`, `blocked`, `done`. `done` is\n**one-way** (no rollback; regression becomes a new feature) per\n[principle 27](https:\u002F\u002Fgithub.com\u002FAnastasiyaW\u002Fcodex-claude-code-config\u002Fblob\u002Fmain\u002Fprinciples\u002F27-feature-tracking.md).\n\n### Mapping between the two\n\n| Doc Status | feature_list.json status | Notes |\n|------------|--------------------------|-------|\n| design | not-started | newly created, no plan yet |\n| planning | in-progress | plan being written |\n| executing | in-progress | code being written |\n| reviewing | in-progress | review\u002Fverify phase |\n| blocked | blocked | identical |\n| done | done | identical |\n\nFor a project that uses this convention, create the doc with `Status: design`\nand the native JSON record with `status: \"not-started\"`. Subsequent transitions\nare manual and follow that project's coordination rule.\n\n## Gotchas\n\n- **ID namespace is project-defined.** It may be global, per layer, or managed\n  by a registry. Discover it before allocation; never treat `F-NNN` as a\n  universal format.\n- **Concurrent allocation.** Scanning and then writing is a race. Use the\n  existing project coordinator or reserve the logical ID with the Git-ref\n  helper before a document\u002Fregistry mutation. A same-ID different-slug\u002Flayer\n  claimant must conflict; re-inventory it, never overwrite it or guess that\n  the durable reservation is stale.\n- **Migration of in-flight features.** When migrating an existing\n  feature into the new format, pass its project-native `--id` explicitly so the\n  feature retains its prior ID in any links from PROBLEMS.md or\n  handoffs. The skill will not auto-detect existing IDs.\n- **Cyrillic titles + Windows.** Per global rule\n  `api-utf8-posting.md`, when writing the markdown file or\n  feature_list.json, always specify `encoding=\"utf-8\"` explicitly to\n  avoid mojibake on Windows.\n- **Layer README table edit.** Preserve the project's existing columns and\n  ordering. If the index cannot be interpreted safely, leave it unchanged and\n  report the unindexed narrative rather than inventing a canonical 5-column\n  row.\n\n## Troubleshooting\n\n| Symptom | Cause | Fix |\n|---------|-------|-----|\n| \"Layer does not exist\" | `docs\u002Flayers\u002F\u003Clayer>\u002F` missing | Run `\u002Flayer-new \u003Clayer>` first |\n| ID reservation conflict | Another request owns that logical ID | Re-inventory project state and use the next native candidate; never overwrite or delete the durable reservation |\n| `feature_list.json` parse\u002Fschema error | Registry is invalid or uses an unknown convention | Preserve the narrative and do not write the registry; report the exact parser\u002Fschema failure and use a project-native adapter when its fields are established |\n| Template missing on this machine | Different host \u002F fresh clone | Pull from public repo: `gh api repos\u002FAnastasiyaW\u002Fcodex-claude-code-config\u002Fcontents\u002Ftemplates\u002Fkb-skeleton\u002Fdocs\u002Flayers\u002F_LAYER-TEMPLATE\u002Ffeatures\u002F_FEATURE-TEMPLATE.md` |\n| Cyrillic in title shows as `?????` | File written without explicit utf-8 | Re-write the file with `encoding=\"utf-8\"`; see `~\u002F.claude\u002Frules\u002Fapi-utf8-posting.md` |\n\n## Implementation note\n\nThis is a **scaffolding** skill: file copy + placeholder replacement +\nsmall JSON merge. Keep it deterministic. The Design \u002F Plan \u002F Verify\nsections of the produced document are meant for the user (or the\nsession that invoked the skill) to fill -- this skill does not\nattempt to generate Design content from the title.\n\nID discovery reads only the sources that the project's convention declares.\nReserve the resulting logical ID atomically, then re-read mutable sources;\ncache is not authority across a concurrent mutation.\n",{"name":5,"model":23,"description":24,"user-invocable":25},"sonnet","Scaffold a feature narrative in an existing project layer with Design, Plan, Verify, and Conclusion sections. Use when: \"create a new feature\", \"start work on feature\", \"scaffold feature doc\", \"\u002Ffeature-new\", or \"begin feature narrative\". Inspects and preserves the target project's feature ID, registry, filename, and coordination conventions; does not create a missing layer (use \u002Flayer-new).\n",true,"main","acf75aba86a23da299be8c69732af3e9b12433cb","upstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Farchitecture\u002Ffeature-new",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":34,"prompt":44,"zipAvailable":25},[35,38,41],{"tool":36,"command":37},"npx","npx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill feature-new",{"tool":39,"command":40},"bunx","bunx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill feature-new",{"tool":42,"command":43},"pnpm","pnpm dlx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill feature-new","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Farchitecture\u002Ffeature-new\u002FSKILL.md",[46],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"scripts\u002Freserve_feature_id.py","SCRIPT","脚本",3940,"https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Farchitecture\u002Ffeature-new\u002Fscripts\u002Freserve_feature_id.py",[53,59,66,73,80,87],{"skillNo":54,"name":55,"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":56,"categories":57,"keywords":58,"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":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"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":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"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":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"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":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"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":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"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",[],[]]