knowledge-explorer
Skill1 starsChanged 5 months agoMITView on GitHub
Systematic knowledge exploration — collect materials, build knowledge trees, visualize, sync to note-taking apps. Trigger words: explore [topic], learn [topic], research [topic], investigate [topic], help me understand [topic], systematically learn [topic]
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Jamie-BitFlight/claude_skillstheir own version, first seen Sep 23, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- Jamie-BitFlight/claude_skills★ 67same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- ai-cognitive-loadSkill
- android-controlSkill
- autonomous-researchSkill
- cron-task-specSkill
- git-worktreeSkill
- hermes-pr-guidelinesSkill
- jj-workspaceSkill
- xurl-best-practicesSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
When the user says:
- Very specific: line 136
Solution: If compatibility issues arise, replace with equivalent graph LR diagram
- Very specific: line 159
If knowledge/ does not exist, create it automatically
- Distinctive: line 160
Ensure the path is writable
- Distinctive: line 288
Must verify upon completion:
- Distinctive: line 326
Check common note-taking app paths
- Very specific: line 327
If none exist, use ~/.openclaw/workspace/
- Distinctive: line 335
Action: Ask the user for a specific direction first
- Distinctive: line 344
Action: Use the most authoritative source available and note it in the document
- Distinctive: line 435
Quick start: must be actionable
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Knowledge Explorer Skill
Version: v1.1
Compatible with: OpenClaw, Claude Code, Claude Desktop
Features: Cross-platform, universal paths, easy configuration
You are a knowledge exploration expert, skilled at quickly mastering new domains and organizing them systematically.
Trigger Conditions
When the user says:
- "explore [topic]"
- "learn [topic]"
- "research [topic]"
- "help me understand [topic]"
- "systematically learn [topic]"
Configuration
Output Path Configuration
Default behavior:
- Prioritize saving to note-taking apps (Obsidian/Notion/Logseq, etc.)
- Backup to OpenClaw workspace
Path priority (auto-detection):
# 1. Obsidian (if detected)
~/Library/Mobile Documents/iCloud~md~obsidian/Documents/Obsidian/knowledge/
# or Linux/Windows Obsidian path
# 2. OpenClaw workspace (default)
~/.openclaw/workspace/
# 3. User-defined path
# Specify in HEARTBEAT.md or in conversationCustom path:
User: explore React, save to ~/Documents/notes/Execution Flow (must follow)
Phase 1: Material Collection (5–10 minutes)
- Search official sources
- Official documentation
- API documentation
- GitHub repositories
- Authoritative tutorials
- Use tools
web_search - search "[topic] official documentation" web_search - search "[topic] tutorial best practices" web_fetch - fetch official documentation content
Phase 2: Build Knowledge Tree (core step)
Strictly follow the 80-15-5 rule:
| Level | Value Share | Content Type | Suggested Count |
| 🌳 Trunk | 80% | Core concepts, must-know skills | 5–10 |
| 🌿 Branch | 15% | Advanced applications, best practices | 5–10 |
| 🍃 Leaf | 5% | Detail configurations, edge cases | 5–10 |
Judgment criteria:
- Trunk: Content you cannot use the tool without understanding
- Branch: Content that significantly improves proficiency once mastered
- Leaf: Details only needed for specific scenarios
Phase 3: Visualize Core Concepts
Use Mermaid to draw 2–4 diagrams (priority from high to low):
- Concept relationship diagram (required)
graph TD Core[Core Concept] --> Concept1[Concept 1] Core --> Concept2[Concept 2] - Learning path diagram (recommended)
graph LR Start[Start] --> Stage1[Phase 1: Basics] Stage1 --> Stage2[Phase 2: Advanced] - Architecture / sequence diagram (if applicable)
graph TB or sequenceDiagram - Timeline / flowchart (optional)
timeline or graph LR
Mermaid chart types (by recommendation):
| Chart Type | Use Case | Recommendation | Stability |
graph TD/LR/TB | Concept relationships, flows | ⭐⭐⭐⭐⭐ | ✅ Stable |
mindmap | Mind maps, knowledge structure | ⭐⭐⭐⭐⭐ | ✅ Stable |
sequenceDiagram | Sequences, interaction flows | ⭐⭐⭐⭐ | ✅ Stable |
erDiagram | Entity relationships, data models | ⭐⭐⭐⭐ | ✅ Stable |
pie | Proportions, distributions | ⭐⭐⭐ | ✅ Stable |
timeline | Timelines, evolution history | ⭐⭐⭐ | ⚠️ Experimental |
gantt | Project plans, schedules | ⭐⭐ | ⚠️ Experimental |
journey | User journeys, storylines | ⭐⭐ | ⚠️ Experimental |
Usage recommendations:
- ✅ Prioritize stable chart types (graph, mindmap, sequenceDiagram, erDiagram, pie)
- ✅ Experimental features can also be used (timeline, gantt, journey)
- ✅ Test compatibility in Obsidian/GitHub/Mermaid Live
- ✅ If rendering issues occur, provide alternatives (e.g., use graph LR instead of timeline)
- ✅ Use standard characters in node names for better compatibility
Notes on experimental features:
timeline: May not be supported in some older Mermaid versionsgantt: Requires specific formatting; rendering is complex- Solution: If compatibility issues arise, replace with equivalent
graph LRdiagram
Phase 4: Output Delivery
Filename format: [Topic]_Knowledge_Tree.md
Save strategy (automatic):
- Detect note-taking app:
# macOS - Obsidian if [ -d ~/Library/Mobile\ Documents/iCloud~md~obsidian/Documents/Obsidian ]; then SAVE_PATH=~/Library/Mobile\ Documents/iCloud~md~obsidian/Documents/Obsidian/knowledge/ # Linux - Obsidian elif [ -d ~/.config/obsidian ]; then SAVE_PATH=~/.config/obsidian/knowledge/ # Fallback: OpenClaw workspace else SAVE_PATH=~/.openclaw/workspace/ fi - Auto-create directory:
- If
knowledge/does not exist, create it automatically - Ensure the path is writable
- If
- Save confirmation:
✅ Knowledge tree saved 📁 [save path]/[Topic]_Knowledge_Tree.md 📱 [sync status, if applicable] 💎 [viewing suggestion]
User-specified path:
User: explore Python, save to ~/Documents/learning/
→ Use user-specified pathFile Structure Template
# [Topic] Knowledge Tree
> Generated: YYYY-MM-DD
> Source: [official documentation link]
> Purpose: [specific learning objective]
---
## 🌳 Trunk: Core Knowledge (80% Value)
### 1. [Core Concept 1]
**Definition**: [concise definition]
**Why it matters**:
- [Reason 1]
- [Reason 2]
**How to use**:
```bash
# Example code or command2. [Core Concept 2]
...
🌿 Branch: Important Knowledge (15% Value)
11. [Advanced Topic 1]
...
🍃 Leaf: Detail Knowledge (5% Value)
17-25. [Detail Configuration]
...
🎯 Learning Path
Phase 1: Master the Core (2–3 hours)
- Learn [Core Concept 1]
- Learn [Core Concept 2]
- Complete quick start
Phase 2: Advanced Practice (3–5 hours)
- Learn [Advanced Topic 1]
- Apply in a real project
Phase 3: Deep Customization (optional)
- Learn [Detail Configuration]
- Performance optimization
💡 Quick Start (5–10 minutes)
Step 1: Installation (X minutes)
# Generic installation commandStep 2: Initialization (Y minutes)
# Initialization commandStep 3: Basic Usage (Z minutes)
# Basic command example❓ FAQ
[FAQ content — decide whether to generate based on Phase 3]
📚 Extended Resources
- Official Documentation: [link]
- Best Practices: [link]
- Community: [link]
- Recommended Tutorials: [link]
🎨 Concept Relationship Diagram
[Mermaid diagram]
🚀 Learning Path Diagram
[Mermaid diagram]
---
## Quality Checklist
Must verify upon completion:
- [ ] Trunk content covers 80% core value
- [ ] At least 2 Mermaid diagrams
- [ ] Includes quick start guide (completable in 5–10 minutes)
- [ ] Provides learning path suggestions
- [ ] File saved to appropriate location
- [ ] Uses universal paths (no hardcoded usernames)
---
## Cross-Platform Support
### macOS
```bash
# Obsidian iCloud
~/Library/Mobile Documents/iCloud~md~obsidian/Documents/Obsidian/knowledge/
# Local Obsidian
~/Documents/Obsidian/knowledge/Linux
# Obsidian
~/.config/obsidian/knowledge/
# or
~/Documents/Obsidian/knowledge/Windows (WSL)
# Obsidian
/mnt/c/Users/$USER/Documents/Obsidian/knowledge/
# or use Windows path notation
C:\Users\[username]\Documents\Obsidian\knowledge\Auto-detection Logic
- Check common note-taking app paths
- If none exist, use
~/.openclaw/workspace/ - User can specify a custom path via conversation
Special Scenario Handling
Scenario 1: Topic Too Broad
Action: Ask the user for a specific direction first
"[Topic] is quite broad. I suggest focusing on:
1. Specific direction A
2. Specific direction B
Which one would you like to learn?"Scenario 2: Official Documentation Not Found
Action: Use the most authoritative source available and note it in the document
Scenario 3: User Specifies a Save Path
Action: Prioritize the user-specified path
Scenario 4: Path Does Not Exist
Action: Auto-create the directory
mkdir -p "[target path]"Scenario 5: Permission Issues
Action: Prompt the user to check permissions and use a fallback path
Universal Design Principles
✅ Do
- Use universal path notation
- ✅
~/(user home directory) - ✅
$HOME/(environment variable) - ✅
~/.openclaw/(standard OpenClaw path)
- ✅
- Avoid hardcoding
- ✅ Auto-detect user environment
- ✅ Provide configuration options
- ✅ Use relative paths
- Cross-platform compatibility
- ✅ Support macOS/Linux/Windows
- ✅ Auto-adapt path formats
- ✅ Provide platform-specific notes
❌ Don't
- Hardcode usernames
- ❌
/Users/username/... - ✅
~/...or$HOME/...
- ❌
- Hardcode absolute paths
- ❌
/home/username/... - ✅ Auto-detect or ask the user
- ❌
- Assume a specific environment
- ❌ Assume Obsidian is used
- ✅ Support multiple note-taking apps
Example Dialogue
User: explore React 18
Assistant:
📚 Phase 1: Material Collection
✅ Found React official documentation
✅ Collected core concepts...
🌳 Phase 2: Build Knowledge Tree
Identified core concepts (80% value):
1. Concurrent Rendering
2. Suspense
3. Transitions
...
🎨 Phase 3: Visualization
✅ Architecture diagram
✅ Learning path diagram
💾 Phase 4: Output
Detected Obsidian...
✅ Saved to Obsidian
📁 knowledge/React_18_Knowledge_Tree.md
📱 iCloud sync enabled
💎 View in Obsidian
🎯 Quick Start (5 minutes):
1. Upgrade to React 18
2. Use createRoot
3. Try useTransitionKey Points to Remember
- 80-15-5 principle is the core
- Key point extraction: definition, importance, usage, pitfalls
- Quick start must be actionable
- Visualization enhances understanding
- FAQ judgment: complex concepts, common misconceptions, high-frequency interview topics → needed
- Universal paths for easy sharing
- Auto-detection reduces configuration
Adoption Guide
How Other Users Can Use This
- Direct use (no configuration needed):
explore [topic]- Auto-saves to
~/.openclaw/workspace/
- Auto-saves to
- Configure Obsidian (optional):
- Install Obsidian
- Enable iCloud sync (macOS)
- Skill auto-detects and uses it
- Custom path:
explore [topic], save to [custom path]
Use this Skill to make knowledge exploration simple, systematic, and shareable! 🚀✨
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

