Neura Market
AgentLint logo

AgentLint

Free

33 evidence-backed checks for AI-friendly repos — file structure, instruction quality, build setup, session continuity, security posture. Claude Code plugin.

FreeFree tier
Type
Open Source
Visit Website GitHub (38 stars)

About AgentLint

AgentLint is a linter for AI agent harnesses — the configuration files, instructions, and setup that AI coding agents read. It scores repositories across 51 deterministic checks on six core dimensions: findability, instructions, workability, safety, continuity, and harness. Additionally, it offers seven opt-in extended checks using AI sub-agents and local session logs. The tool works with Claude Code, Codex, and Cursor, providing an interactive scan-fix-report flow via the /al command or CLI. It is open source, evidence-backed, and aims to reduce AI slop by ensuring the harness is correct.

Key Features

51 deterministic checks across 6 core dimensions (findability, instructions, workability, safety, continuity, harness)
7 opt-in extended checks using AI sub-agents and local Claude Code session logs
Interactive scan-fix-report flow via the /al slash command
Generates HTML report after auto-fixing issues
Works with Claude Code, Codex, and Cursor AI coding agents
CLI tool that operates on any git repository
Installable via npm (npm install -g agentlint-ai) with optional Claude Code plugin registration
Evidence-backed, zero-opinion check logic

Pros & Cons

Pros
  • Comprehensive check suite covering multiple critical dimensions
  • Works with multiple popular AI coding agents (Claude Code, Codex, Cursor)
  • Provides actionable fixes and re-scoring with HTML report generation
  • Open source and free to use
  • Evidence-backed rather than opinionated

Best For

Auditing and improving CLAUDE.md and AGENTS.md files for AI agent clarityEnsuring repository configuration (gitignore, CI, hooks) is AI-friendlySecuring projects against secrets exposure by AI agentsOptimizing build and setup instructions for agent reproducibilityStandardizing harness quality across teams using AI coding agents