How easy is your codebase for AI agents to work in?
Paste a GitHub link. GoodRepo checks how the code is organized, what instructions it leaves for AI tools, and how easy it is to test, then gives it a score out of 100. 40 checks, each with the reason behind it.
Scan from inside your coding agent
npx skills add jtljrdn/goodrepoWhat gets scored
Discoverability
Can an agent quickly find the code that matters?
“Source lives under a single predictable root”6 checksInstructions
Are agent-facing instructions present and complete?
“AGENTS.md exists”9 checksTestability
Can an agent verify its own changes?
“typecheck script defined”7 checksConsistency
Does the codebase repeat the same patterns?
“One validation approach across routes”6 checksTooling
Is setup, build, lint and test behavior obvious?
“.env.example lists required variables”8 checksContext Efficiency
How much must an agent read to make one change?
“Common changes stay inside one module”4 checks