メニュー
General extraction/ingestion skill that routes to specific workflows based on input type. Extracts structured information from documents, emails, reviews, feedback, and other sources.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/nicsuzor/aops --skill extractコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Core academicOps skill — institutional memory, strategic coordination, workflow routing, and framework governance. Merges butler (chief-of-staff) with framework development conventions.
Daily note lifecycle — compose and maintain a factual daily note. Reports the state of the day; does not prioritise or recommend. SSoT for daily note structure.
Emergency session bail — fast resume task + short handover, no commit/PR/reflection. For when you (or the user) need a clean context now. Use /end-session for canonical close.
SOC 職業分類に基づく
SKILL.md を表示中
| name | extract |
| type | skill |
| category | instruction |
| description | General extraction/ingestion skill that routes to specific workflows based on input type. Extracts structured information from documents, emails, reviews, feedback, and other sources. |
| triggers | ["extract information","extract from document","ingest","extract decisions","extract training data","convert document","DOCX/PDF/XLSX conversion","/convert-to-md"] |
| modifies_files | true |
| needs_task | true |
| mode | execution |
| domain | ["operations"] |
| allowed-tools | Read,Grep,Glob,Edit,Write,Skill,Bash |
| version | 1.0.0 |
| permalink | skills-extract |
Taxonomy note: This skill provides domain expertise (HOW) for extracting structured information from documents and sources. See [[aops-core/skills/remember/references/TAXONOMY.md]] for the skill/workflow distinction.
General-purpose extraction skill that intelligently routes to specialized workflows based on input type. Extracts structured information from various sources and stores it appropriately (public framework vs. private data).
Universal axioms apply (enforced by rbg). P#52 (Read-Then-Write Memory) is especially relevant.
MANDATORY before creating any new extracted content: search PKB for existing knowledge on the same subject.
mcp__pkb__search(query="[topic/person/document subject]")
This prevents duplicate memories and grounds extraction in accumulated knowledge. See [[remember]] skill's "search first" step as the model.
Provide a unified entry point for all extraction tasks:
When invoked, analyze the input and route to the appropriate workflow:
Signals:
Route to: workflows/training-data.md
Storage:
$ACA_DATA/processed/review_training/Signals:
Route to: Archive extraction logic (selective extraction, use /remember for storage)
Storage: Use Skill(skill="remember") for PKB storage
Signals:
Route to: Existing aops-core/skills/decision-extract/SKILL.md
Storage: Daily note with decision formatting
Signals:
Route to: workflows/document-knowledge.md (to be created)
Storage: Depends on content - PKB or framework docs
Signals:
/convert-to-md (alias preserved for backwards compatibility)Route to: workflows/docs-to-md.md
Storage: Converted .md files replace originals in the same directory
procedures/review-inline-comments.mdreview.txt + source.pdf + metadata.jsonworkflows/revision-history.md (to be created)See procedures/review-inline-comments.md for detailed procedure.
Quick summary:
CRITICAL: Training data often contains sensitive information (author names, unpublished work, specific critiques).
Sensitive data → $ACA_DATA/processed/review_training/{collection_name}/:
extracted_examples.json (full text/feedback pairs)training_pairs.jsonl (machine-readable format)collection_summary.md (with identifying information)Generalized patterns → Framework (public repo):
aops-core/skills/hydrator/workflows/peer-review.md (update with principles)aops-core/skills/*/references/ (depersonalized examples)High-quality extraction:
Generalization quality:
Apply selective extraction logic.
Key principle: Most archival documents have NO long-term value. Be highly selective.
Extract: Concrete outcomes, significant relationships, financial records Skip: Newsletters, invitations, administrative routine, mass communications
Storage: Use Skill(skill="remember") with proper tags and canonical identifiers.
Delegate to aops-core/skills/decision-extract/SKILL.md.
Key principle: Extract tasks requiring approval/choice that are blocking other work.
Storage: Daily note with formatted decision list for batch processing.
$ACA_DATA/processed/Directory structure:
$ACA_DATA/processed/
├── review_training/
│ ├── {collection_name}/
│ │ ├── extracted_examples.json
│ │ ├── training_pairs.jsonl
│ │ ├── collection_summary.md
│ │ └── source_documents/
│ └── ...
├── email_archive/
│ └── ...
└── ...
Access: This directory is:
/home/nic/brain)When adding examples to public framework docs:
/extract vs. specialized skillsUse /extract:
Use specialized skill directly:
/remember - When you know you want to add to knowledge base/decision-extract - When specifically extracting decisions/review-training - When processing matched review/source pairs (legacy)Note on /convert-to-md: This trigger is now an alias for /extract. Invoking /convert-to-md routes to the workflows/docs-to-md.md workflow.
/extract → analyze input → route to:
- /remember (for archival preservation)
- /decision-extract (for pending decisions)
- training-data workflow (for LLM training data)
- document-knowledge workflow (for general extraction)
| Scenario | Behavior |
|---|---|
| Unclear input type | Ask user to clarify extraction goal |
| Cannot convert document format | Try alternative conversion, document failure |
| Ambiguous feedback | Flag with "quality": "ambiguous", include with caveats |
| No clear extraction value | Ask user if they want to skip or force extraction |
| Storage location unclear | Default to $ACA_DATA/processed/, confirm with user |
Input: DOCX file with inline comments from peer review
User: /extract /path/to/review.docx --type peer-review
Agent:
pandoc --track-changes=all$ACA_DATA/processed/review_training/aoir2026/aops-core/skills/hydrator/workflows/peer-review.md with depersonalized principlesInput: Directory of email MSG files
User: /extract emails/2025-Q1/ --type archive
Agent:
/remember to store in PKBInput: Mixed documents without type specified
User: /extract documents/
Agent:
Before completing extraction:
Completeness:
Quality:
Sensitivity:
$ACA_DATA/processed/Documentation: