Lintrule
Lintrule is a command-line tool that utilizes large language models (LLMs) for conducting code reviews. It enables software developers to enforce policies that ...
Last verified:
What is Lintrule?
Lintrule is a command-line tool that leverages large language models to automate code reviews, enforcing complex policies beyond traditional linters and catching bugs that tests might miss. Developers write simple rules in plain Markdown files stored in a .rules folder, specifying instructions like preventing customer data logging to maintain SOC2 compliance. It runs efficiently on git diffs, checking only changed code by default, such as the last commit or specified branches, and provides clear pass/fail results with explanations for failures.
Key features include easy installation via a curl script on MacOS, Linux, and WSL, parallel processing of multiple files and rules for consistent speed, and seamless integration with CI/CD pipelines like GitHub Actions using a secret key. Rules support frontmatter for file-specific includes, such as targeting only SQL files for database migration checks. It supports any programming language since it uses LLMs, making it versatile for diverse codebases from small projects to large ones like Rust or Next.js.
Lintrule is ideal for software development teams seeking to reduce manual review burdens, prevent outages from subtle issues, and maintain high code quality without wasting time on basic 'LGTM' approvals. It's particularly useful for enforcing compliance, reviewing performance-sensitive changes, and scaling reviews across contributors without proportional cost increases.
Lintrule pricing
Pricing model: Paid
Lintrule pros
- Runs on git diffs to check only changed code
- Rules written in plain Markdown language
- Supports any programming language via LLMs
- Parallel processing for consistent fast checks
- Easy CLI install with curl script
- Integrates with GitHub Actions seamlessly
- File-specific rules using frontmatter includes
- Cost estimator command available
- Reduces manual human review time
- Enforces complex policies like SOC2 compliance
- Catches bugs missed by traditional tests
- Non-flaky false positives when rules are specific
- Smart diff detection in CI environments
- Scales to large projects like Rust repo
- Free to get started, no credit card needed
- Provides detailed failure explanations
- Supports custom diffs like main..feature-branch
Lintrule cons
- Usage-based pricing at $1 per 1k changed lines
- Potential false positives with vague rules
- Requires internet for LLM cloud processing
- No explicit free tier beyond trial
- Costs scale with code volume and contributors
- Needs secret management for CI integration
- Dependent on LLM quality and pricing trends
- Fetch-depth 2 required in GitHub checkout
- Login required after init for authentication