General-purpose coding policy for Baruch's AI agents
92
91%
Does it follow best practices?
Impact
93%
1.27xAverage score across 10 eval scenarios
Advisory
Suggest reviewing before use
rules/ci-safety.md's "Never Skip Tests" and rules/context-artifacts.md's "Disagreeing With the Reviewer"commit-conventionsrules/commit-conventions.md's "Keep PRs focused" and this rule appear to conflict. They don't. Focus governs the SHAPE of the bundle (one logical change per commit / PR); boy-scout governs whether you walk away from problems you noticed (no, you don't)evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
rules
skills
install-reviewer