LaunchKit · 2026
Back to Skills

frankenstein

Combine the best parts of multiple skills into one.

4
1.1k downloads
by @rubenaquispe

Setup & Installation

openclaw skills install @rubenaquispe/frankenstein

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

npx clawhub install frankenstein

Version History

v1.2.0Feb 7, 2026 - Version note by skill developer:

Search all sources: ClawHub, GitHub, skills.sh, skillsmp.com, and other AI skill repos

What This Skill Does

Searches multiple AI skill repositories for matching skills, security-scans each candidate, and synthesizes a new skill combining the strongest features from each source. Runs a multi-pass vetting loop before presenting the combined result for human review.

Automates the tedious work of manually evaluating, security-scanning, and cherry-picking across competing skills from multiple repositories.

When to use it

  • Merging five competing SEO audit skills into one comprehensive tool
  • Building a unified code review skill from fragmented open-source options
  • Consolidating overlapping web scraping skills into a single workflow
  • Selecting the best-implemented feature across multiple data pipeline skills
  • Replacing three separate testing skills with one that covers all their strengths

Example Workflow

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

INPUT

User asks: Frankenstein me an SEO audit skill

AGENT
  1. 1Search ClawHub, GitHub, skills.sh, and skillsmp.com for SEO audit skills
  2. 2Run each candidate through skill-auditor and skip any with a risk score below 7
  3. 3Analyze safe skills in sandwrap read-only mode to extract features, methodology, and unique strengths
  4. 4Build a comparison matrix and select the winning approach for each feature
  5. 5Assemble the combined skill with skill-creator, run three vetting passes, then present for human approval
OUTPUT

A new SKILL.md combining the best methodology, rules engine, and auto-fix capabilities from safe candidates, with full source attribution