trace-companies

Trace company affiliation for repositories via mapping, GitHub API, and web search.

Updated Apr 7, 2026
One-click install
npx skills add https://github.com/noemotiovon/oss-x --skill trace-companies
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trace-companies
Source: https://github.com/noemotiovon/oss-x/tree/main/.claude/skills/trace-companies
Command: npx skills add https://github.com/noemotiovon/oss-x --skill trace-companies

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Trace company affiliation for each repository using a three-layer approach: known static mapping, GitHub API attribution, and LLM web search for unresolved cases.

Core Features & Use Cases

  • Layered resolution: resolve affiliation via mapping, API, and web search with human review for unknowns.
  • Batch efficiency: processes large repo lists produced by the OSS-X pipeline and outputs per-repo company data.
  • Evidence-based: captures evidence URLs and confidence levels to justify each assignment for auditing.

Quick Start

Run python3 scripts/trace_companies.py output/repo_exp.csv -o output/company.csv --summary to generate the company mapping and summary report.

Frequently Asked Questions about trace-companies

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I trace company affiliations for a list of GitHub repositories?

Trace company affiliations by applying a three-layer approach using known static mapping, GitHub API attribution, and LLM web search to resolve each repository's owner and output a structured CSV file.

What is the best way to find the corporate owner of unresolved GitHub repositories?

Find corporate owners for unresolved repositories by using LLM web search as a fallback layer, which researches unknown cases and records evidence URLs alongside confidence levels for auditing.

How to map expanded repository lists to their parent companies in batch?

Map expanded repository lists to parent companies in batch by processing the input CSV through layered resolution, optimizing large workloads and generating a summary report of company attributions.

Does tracing repository affiliations support human review for low confidence results?

Tracing repository affiliations supports human review by flagging unresolved cases, capturing evidence URLs, and assigning confidence levels so you can audit assignments before finalizing the CSV.

Can I use the GitHub API to resolve company names for open source projects?

Use the GitHub API to resolve company names by looking up user and organization profiles as the second resolution layer, catching cases that static mapping misses before invoking web search.

Why do I need evidence URLs and confidence levels when attributing repos to companies?

Evidence URLs and confidence levels are needed when attributing repos to justify each assignment for auditing, ensuring downstream analysis relies on verifiable data rather than guesses.