Evaluate AGENTS.md instruction files via controlled A/B testing.
npx skills add https://github.com/vltansky/skills --skill agents-md-evals
Please help me install this Agent Skill. Skill: agents-md-evals Source: https://github.com/vltansky/skills/tree/main/skills/agents-md-evals Command: npx skills add https://github.com/vltansky/skills --skill agents-md-evals
This Skill helps you identify and remove redundant or ineffective instructions from your AI configuration files (like AGENTS.md or CLAUDE.md), saving valuable context tokens and improving AI performance.
Use the agents-md-evals skill to evaluate your AGENTS.md file.
High-intent search queries and answers about installing and using this skill.
Benchmark AI instruction performance by grading rules against codebase patterns to pinpoint which instructions do not earn their tokens, allowing you to remove ineffective rules and improve AI performance.
Clean file isolation protocols are required for baseline runs to ensure accurate A/B testing. You also need git log analysis for prompt generation to properly evaluate and benchmark instruction performance.
Yes, the Skill evaluates extensive configuration files by rigorously testing instruction rules, categorizing them by coding style or workflow, and benchmarking performance to show exactly which rules are essential.
Evaluating AI configuration files requires git log analysis for prompt generation and clean file isolation protocols for baseline runs to ensure accurate A/B testing and performance benchmarking.
LISANG3Automates Git worktree creation for isolated development with setup and tests.
borkwebAudit and trim AI agent instruction files to remove redundant documentation.
hschneCreates standardized AI skills with validation, eval testing, and optimization workflows.
BoraPerusicEvaluate LLM outputs with automated metrics and human judgments.
