LaunchKit · 2026
Back to Skills

pattern-finder

Discover what two sources agree on — find the signal.

3
1.3k downloads
by @leegitw

Setup & Installation

openclaw skills install @leegitw/pattern-finder

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install pattern-finder

Version History

v1.0.1Feb 10, 2026 - Version note by skill developer:

Migrated to public GitHub repo, updated homepage URLs

What This Skill Does

Compares two text sources to find shared principles — ideas that appear in both, even when expressed differently. Uses an N-count system where N=2 means two independent sources agree on the same idea. Shows what's shared, what's unique to each source, and flags divergence types.

Independent sources converging on the same principle is stronger evidence than a single source, and this skill surfaces that convergence automatically instead of requiring manual side-by-side review.

When to use it

  • Finding where two research papers reach the same conclusion
  • Checking if two coding philosophy guides share the same principles
  • Validating a personal framework against a published methodology
  • Comparing notes from two different mentors for overlapping advice
  • Cross-checking two extracted knowledge bases for redundancy

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: 'Do these two extracted principle sets agree on anything?'

AGENT
  1. 1Receives two sources (raw text, extractions, or mixed)
  2. 2Identifies the core principles in each source
  3. 3Searches for semantic matches, not just keyword overlap
  4. 4Categorizes results into shared, unique-to-A, and unique-to-B
  5. 5Assigns N-count and confidence level to each shared pattern
OUTPUT

A structured breakdown listing N=2 validated shared patterns, unique principles per source, divergence analysis, and suggested next steps such as adding a third source to reach N≥3