메뉴
Use when user says "generate report about X", "create summary of X", "stwórz podsumowanie X". Generates a requested format (summary, reading list, topic map, timeline) and saves to `_outputs/` or topic folder.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | output |
| description | Use when user says "generate report about X", "create summary of X", "stwórz podsumowanie X". Generates a requested format (summary, reading list, topic map, timeline) and saves to `_outputs/` or topic folder. |
Trigger phrases: "generate report about X", "create summary of X", "stwórz podsumowanie X". The user wants a derived artifact built from vault content — not a full new wiki article (that is COMPILE), but a focused report or summary.
vault-map.md → catalog.md → graph.md.content/_outputs/<format>/YYYY-MM-DD_<topic>.md.type:.CLAUDE.md "Navigation Protocol" — read on every operation before this workflow.
compile skill — for full wiki articles (different output type).Use when user says "export [[Note]] ...", "export
Use when user says "import <file.zip>", "zaimportuj paczkę", "import bundle/pack of notes". Imports a brain-pack zip (made by /export in another brain) into this vault - validates, triages collisions, places notes into this vault's taxonomy, stamps provenance, updates all 3 indexes, writes a report. At most 2 user prompts per run.
Use when user says "ingest", "process inbox", "przetworz nowe pliki", or when files appear in `content/_raw/inbox/`. Processes raw sources into wiki notes, handles intra-batch clustering with confirmation, updates all 3 indexes.
Use when user says "lint", "health check", "audit". Checks vault for missing frontmatter, broken wikilinks, orphans, stub notes, inconsistent tags, TODO markers, stale content; saves report to `_outputs/reports/`.
Use when user says "research X", "what do my notes say about X", "co mam w notatkach o X". Synthesizes an answer from the vault citing wikilinks; offers to save substantial answers to `_outputs/answers/`.
Read research outline, launch independent agent for each item for deep research. Disable task output.