用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/Abilityai/cornelius --skill analyze-kb命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | analyze-kb |
| description | Analyze knowledge base structure and update the knowledge-base-analysis.md report |
| automation | gated |
| allowed-tools | Read, Bash, Glob, Grep |
Analyze the Obsidian knowledge base structure and regenerate a condensed, manageable analysis report.
Use Local Brain Search for all semantic search operations.
Scripts:
# Semantic search
resources/local-brain-search/run_search.sh "query" --limit 10 --json
# Find connections
resources/local-brain-search/run_connections.sh "Note Name" --json
# Get graph statistics
resources/local-brain-search/run_connections.sh --stats --json
# Find hub notes
resources/local-brain-search/run_connections.sh --hubs --json
# Find bridge notes
resources/local-brain-search/run_connections.sh --bridges --json
Follow these steps:
Get vault statistics using Local Brain Search:
resources/local-brain-search/run_connections.sh --stats --json
List all notes with full directory tree using Bash find:
find $VAULT_BASE_PATH/Brain -name "*.md" -type f | head -50
Use Glob to explore directory structure:
Brain/**/*.mdBrain/02-Permanent/*.mdBrain/03-MOCs/*.mdBrain/01-Sources/**/*.mdRead key hub notes from major clusters using Read tool (Dopamine, Self/Buddhism, Decision-Making, Flow, etc.)
Use Local Brain Search to identify hub notes and bridges:
resources/local-brain-search/run_connections.sh --hubs --json
resources/local-brain-search/run_connections.sh --bridges --json
Fingerprint scope note: --hubs/--bridges are the cognitive-fingerprint
axis. Scope enforcement is now ON (MEMORY_CONFIG['scope']['enforce']=True,
since 2026-06-25; see SCOPE-IMPLEMENTATION-PLAN.md), so they compute on the
core subgraph only (02-Permanent, 03-MOCs, AI Extracted Notes,
01-Sources). Document-Insights-inflated hubs (Dopamine/Flow) no longer appear;
the true core hubs (MOC - Eight-Circuit, Decision Making, Gilbert, Tetlock,
Hagen) surface. --stats stays whole-graph (it answers "how big is the
store", not "who am I"). Note which mode is active when recording hubs.
Pull BDG statistics for layer distribution and lifecycle phases:
resources/brain-graph/run_brain_graph.sh status --json
Analyze thematic clusters, hierarchical organization, and conceptual architecture
Use Edit or Write to update the knowledge-base-analysis.md file with a condensed report (target: ~600-800 lines max):
CRITICAL OUTPUT REQUIREMENTS:
Generate a condensed, scannable, high-signal report that reveals the knowledge base's personality, cognitive fingerprint, and growth opportunities without overwhelming detail.
| Source | Location | Read | Write | Description |
|---|---|---|---|---|
| Brain notes | Brain/**/*.md | X | All vault notes for analysis | |
| Permanent notes | Brain/02-Permanent/ | X | Core insights | |
| MOCs | Brain/03-MOCs/ | X | Map of Content hubs | |
| Sources | Brain/01-Sources/ | X | Reference materials | |
| AI Insights | Brain/AI Extracted Notes/ | X | AI-extracted notes | |
| Document Insights | Brain/Document Insights/ | X | Research session notes | |
| Book scopes | Brain/Books/ | X | Per-book scopes (non-core, family) | |
| Changelogs | Brain/05-Meta/Changelogs/ | X | Session logs | |
| Local Brain Search | resources/local-brain-search/ | X | Graph statistics, hubs, bridges | |
| Knowledge base analysis | knowledge-base-analysis.md | X | X | Output report |