A linter and LSP that validates configuration files for AI coding assistants like Claude Code, Cursor, and Copilot.
agnix is a specialized linter and language server designed to validate configuration files for AI coding assistants. It checks files like CLAUDE.md, SKILL.md, and MCP configs to ensure they are syntactically correct and adhere to tool-specific rules, preventing silent failures where AI tools ignore misconfigured skills or instructions. The tool supports auto-fixes and integrates directly into popular code editors and CI pipelines.
Developers and teams using AI coding assistants like Claude Code, Cursor, GitHub Copilot, or Kiro who want to ensure their agent configurations are correct and reliable.
Developers choose agnix because it provides comprehensive, multi-tool validation with auto-fixes, catching configuration errors that would otherwise cause AI assistants to fail silently. Its extensive rule set and editor integrations make it the most robust solution for maintaining reliable AI coding workflows.
The missing linter and lsp for AI coding assistants. Validate CLAUDE.md, AGENTS.md, SKILL.md, hooks, MCP. Plugin for all major IDEs included, with autofixes.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Validates 385+ rules across Claude Code, Cursor, Copilot, and more, ensuring comprehensive support for multi-tool AI stacks as listed in the README.
Offers auto-fix capabilities with safe/unsafe confidence levels, reducing manual correction effort for common configuration mistakes highlighted in the CLI example.
Provides plugins for VS Code, JetBrains, Neovim, and Zed, enabling real-time linting directly in the development environment as detailed in the editor integration section.
Includes a GitHub Action for automated validation in CI/CD pipelines, facilitating continuous checks on agent configurations without manual intervention.
New rules and tool support ship constantly, which can lead to frequent updates and potential breaking changes requiring user vigilance, as noted in the README's update disclaimer.
While extensive, the rule set may not include all AI assistants or the latest features immediately, leaving gaps for users of newer or less common tools.
The --fix-unsafe mode applies low-confidence fixes that could introduce errors, necessitating careful review after automatic corrections, as warned in the CLI usage.