Fake Star Audit
Audit GitHub repos for inflated star patterns with transparent LOW, MEDIUM, and HIGH ratings and per-rule evidence before you adopt a skill or dependency.
Overview
fake-star-audit is a MCP server for the Idea phase that scores GitHub repos for fake-star risk with LOW, MEDIUM, or HIGH labels and rule-level evidence.
What is this MCP server?
- Rule-based GitHub fake-star detector with explicit LOW, MEDIUM, and HIGH verdicts
- Per-rule evidence so agents and humans can see why a repo was flagged
- Transparent methodology rather than opaque black-box popularity scoring
- Python package fake-star-audit v0.1.1 with uvx and stdio MCP transport
- Fits agent workflows that vet skills.sh listings and MCP servers before install
- Server version 0.1.1
- PyPI identifier fake-star-audit with uvx runtime hint
What problem does it solve?
Trending GitHub stars are easy to game, so solo builders cannot tell if a hyped skill or library is genuinely trusted.
Who is it for?
Builders shortlisting MCP servers, agent skills, or OSS repos where social proof is mostly GitHub stars.
Skip if: Teams that already rely on enterprise SBOM and vendor risk platforms for all dependency decisions.
What do I get? / Deliverables
You get an explainable fake-star risk rating with per-rule evidence before committing to install or fork a repository.
- LOW, MEDIUM, or HIGH fake-star risk classification
- Per-rule evidence breakdown for transparency
- Agent-ready audit summary for adopt-or-skip decisions
Recommended MCP Servers
Journey fit
Discover is the canonical shelf because builders first encounter suspicious GitHub hype while hunting tools, skills, and OSS to try. Discover covers due diligence on trending repos where star count is a primary trust signal solo builders cannot easily verify by hand.
How it compares
GitHub star-integrity audit MCP, not a general vulnerability scanner or license checker.
Common Questions / FAQ
Who is fake-star-audit for?
fake-star-audit is for solo and indie builders who discover tools on GitHub and want rule-based evidence before trusting star counts.
When should I use fake-star-audit?
Use it during discover and scope validation whenever a repo’s popularity is a deciding factor for adopting a skill, MCP server, or library.
How do I add fake-star-audit to my agent?
Install the PyPI package fake-star-audit via uvx, register it as a stdio MCP server in your agent host, then call its audit tools with a GitHub repo identifier.