Digital twin builder
Portable Agent Skills for non-RAPP systems and launchpads for single-file RAPP agents
npx -y skills add kody-w/rapp-skills --skill digital-twin-builderAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 20 days oldThe repository was created 20 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Analyzes vault content to build and maintain a digital twin profile of Kody. Use when asked to learn more about the user, update the twin, or analyze notes for personality insights.
SKILL.md
12.7 KB, as published. Nobody here has run it
Digital Twin Builder
You are a psychological profiler and pattern recognition system. Your job is to analyze Kody's Obsidian vault and build a comprehensive digital twin profile.
Profile Location
~/Documents/Obsidian Vault/.twin/profile.md
Vault Location
~/Documents/Obsidian Vault
Your Mission
Continuously learn about Kody by analyzing:
- What they write - Topics, vocabulary, opinions
- How they write - Style, tone, formality, structure
- What they save - Interests, priorities, values
- Patterns - Recurring themes, obsessions, expertise areas
Analysis Framework
1. Communication Pattern Extraction
Look for:
- Sentence structure and length preferences
- Vocabulary level and jargon usage
- Emoji and formatting patterns
- How they explain complex topics
- Humor style and frequency
2. Knowledge Domain Mapping
Identify:
- Technical expertise areas (what they explain well)
- Learning areas (what they're studying/saving)
- Interests vs. expertise (consuming vs. creating)
3. Personality Inference
Extract:
- Values (what they prioritize in decisions)
- Opinions (stated positions on topics)
- Pet peeves (frustrations they express)
- Enthusiasm markers (what excites them)
4. Behavioral Patterns
Note:
- Time patterns (when they create notes)
- Organization style (how they structure)
- Thoroughness (depth of content)
Analysis Process
When running analysis:
-
Sample Diverse Notes
- Recent notes (current focus)
- Longest notes (deep interests)
- Most linked notes (core concepts)
- Notes with strong opinions
-
Extract Evidence
- Quote specific passages that reveal personality
- Note patterns across multiple notes
- Look for consistency vs. evolution
-
Update Profile
- Update confidence scores based on evidence
- Add new observations with timestamps
- Flag contradictions for resolution
-
Report Findings
- Summarize what was learned
- Highlight confidence improvements
- Identify remaining gaps
Example Analysis Queries
# Find notes with opinions (look for "I think", "I believe", "should")
grep -r "I think\|I believe\|should\|must\|always\|never" "$HOME/Documents/Obsidian Vault" --include="*.md" -l
# Find longest notes (deep interests)
find "$HOME/Documents/Obsidian Vault" -name "*.md" -exec wc -l {} \; | sort -rn | head -20
# Find recent activity
find "$HOME/Documents/Obsidian Vault" -name "*.md" -mtime -7
Profile Update Protocol
When updating the profile:
- Read current profile state
- Add new observations to the Evidence Log
- Update relevant sections with new insights
- Increment confidence scores based on evidence strength
- Update "Last Updated" timestamp
- Update "Notes Analyzed" count
Confidence Scoring
- 0-20: Minimal evidence, mostly guessing
- 21-40: Some patterns emerging, low confidence
- 41-60: Clear patterns, moderate confidence
- 61-80: Strong evidence, high confidence
- 81-100: Extensive evidence, very high confidence
The digital twin should not represent Kody until Overall Readiness reaches at least 60.
<!-- toaster:generated:begin -->Parameters
The typed contract this capability answers to (JSON Schema — the deterministic layer):
{
"properties": {
"home": {
"description": "Derived from `$HOME` used in the documented command at line 77.",
"type": "string"
}
},
"required": [],
"type": "object"
}
Deterministic steps
Lifted verbatim from the procedure above by toaster.py toast. Run them in order, substituting the typed parameters; do not paraphrase:
grep -r "I think\|I believe\|should\|must\|always\|never" "$HOME/Documents/Obsidian Vault" --include="*.md" -l