Skip to main content
Category: GTM Intelligence · Version: 0.0.1 · Author: opendirectory · Compatibility: All targets (claude, codex, opencode, gemini, anti-gravity, openclaw, hermes)
Give this skill an SDK or library name. It searches GitHub for public repos that import it, scores each repo by company signal and activity, identifies who is building on you in production, and outputs a ranked adoption report with outreach context for high-signal company repos.

Install

What It Does

  • Searches GitHub code for all public repos that import or require the SDK
  • Supports npm, Python, Go, Ruby, and generic import patterns — auto-detected from the SDK name
  • Filters forks, tutorials, examples, and archived repos from results
  • Classifies each repo: company org, affiliated developer, solo developer, or tutorial noise
  • Scores by adoption signal: org type, company field, stars, recency, fork status
  • Fetches owner profile and top contributor for each high-signal repo
  • Tracks adoption velocity: new repos in last 7, 14, and 30 days
  • Compares to previous snapshot for exact new-adopter count on repeat runs
  • Generates an outreach brief per high-signal company repo: what they are building, who introduced the SDK, and what to say
  • Saves output to docs/sdk-adopters/[sdk-name]-[date].md

Requirements

Setup

How to Use

Include a short description of your product and the skill will tailor the outreach message to your context.

Why Score Instead of Just List

A raw GitHub code search for a popular SDK returns tutorials, forks, abandoned side projects, and examples mixed in with real production users. The adoption score separates them:
  • A company org repo with recent commits and 200 stars scores 130+
  • A tutorial named “learn-stripe-payments” scores below 20 and is excluded
The person who added the import in a company repo is the warmest possible lead. They chose the SDK, they understand what it does, and they have an active production use case.

The Adoption Score

score = org_signal + company_field + stars_capped + recency + quality_flags
  • Org type (50 points): GitHub Organization accounts are almost always a company or serious team
  • Company field (20 points): user has a company listed on their GitHub profile
  • Stars (up to 50 points): capped at 500 stars to prevent established open-source from dominating
  • Recency (up to 50 points): pushed in last 7 or 30 days
  • Quality (up to 40 points): not a fork, not archived, not a tutorial
Score >= 80: full outreach brief generated. Score 40-79: listed in report. Score < 40: counted in breakdown only.

Velocity Tracking

Run the skill weekly and save the JSON snapshot. On each subsequent run, the skill compares the current repo list to the previous snapshot and tells you exactly how many new teams adopted the SDK since last time.

Cost Per Run

  • GitHub Code Search API: free with token
  • GitHub User, Org, Contributors API: free with token
  • AI analysis: uses the model already running the skill; no additional cost
  • Total: free

Standalone Script

Run the data-fetching step directly from the terminal without Claude. Useful for scheduled jobs or CI pipelines.
The script handles Steps 3-5 (code search, scoring, enrichment) and writes a JSON file. Open that file with Claude and ask: “Generate adoption briefs from this SDK data.”

Project Structure

License

MIT