agentlint

Local-first CLI tool for continuous improvement of AI-assisted development workflows.

Overview

agentlint traces issues to their origins and provides preventive recommendations that compound value over time. Unlike traditional linters, agentlint analyzes the AI development system itself—configuration quality, session effectiveness, and workflow optimization.

Installation

Binary (Recommended)

# macOS / Linux
curl -fsSL https://raw.githubusercontent.com/Obsidian-Owl/agentlint/main/scripts/install.sh | bash

npm

# Requires Node.js 22+
npm install -g @agentlint/cli

From Source

# Requires Bun 1.x
git clone https://github.com/Obsidian-Owl/agentlint.git
cd agentlint
bun install
bun run src/cli.ts --version

Usage

# Show version
agentlint --version

# Show help
agentlint --help

# Update to latest version
agentlint update

Development

# Install dependencies
bun install

# Run tests
bun test

# Lint code
bun run lint

# Format code
bun run format

# Type check
bun run typecheck

# Build
bun run build

Architecture

agentlint follows a 6-layer architecture:

1. CLI - Command-line interface
2. Orchestration - Claude Agent SDK integration
3. Tools - Analysis capabilities
4. ACT Adapters - Agent context tool adapters
5. Persistence - SQLite storage
6. Integration - External service connections

See docs/architecture for detailed documentation.

Temporal Analysis (EP09)

Track workflow effectiveness over time with mixed-methods measurement:

• Quantitative: Baseline capture, delta comparison, trend analysis
• Qualitative: Structured reviews with 6 dimensions and Likert scoring

See src/temporal/README.md and specs/ep09-temporal-analysis/quickstart.md.

License

MIT