open-source, local-first AI CLI for developers. It connects to your local Ollama models or remote providers like OpenAI, Anthropic, Gemini, and Hugging Face, letting you chat with your code, review changes, and craft commits . all from the terminal.
git clone https://github.com/meer-ai/meer.gitMeerAI is an open-source AI CLI that brings intelligent code assistance directly to your terminal without browser tabs or SaaS lock-in. It supports local Ollama models and remote providers including OpenAI, Anthropic, Gemini, and Hugging Face, letting you chat with your codebase, review changes, and generate conventional commits. The tool offers two agent workflows—classic streaming and structured tool-driven—with 60+ built-in tools for file operations, git management, semantic search, testing, and security scans. MeerAI stores conversations and long-term memory locally by default, integrates with Model Context Protocol servers, and provides a rich terminal UI with approval overlays, timelines, and diff previews. Developers can configure multiple provider profiles, override settings per command, and extend functionality with custom slash commands.
["Install `meer` via your package manager (e.g., `brew install meer` or `npm install -g meer-cli`).","Navigate to your project directory in the terminal.","Use `meer` commands like `meer chat` for interactive sessions, `meer review` to analyze code changes, or `meer commit` to generate commit messages from staged changes.","For Git workflows, pipe diffs to `meer` (e.g., `git diff | meer review`).","Customize the AI model by setting `OPENAI_API_KEY` or configuring Ollama locally for faster, offline responses."]
Interactive code review and refactoring via terminal chat
Generate conventional commit messages from staged changes
Repo-aware Q&A and documentation with memory persistence
Semantic code search and project scaffolding
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/meer-ai/meerCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Use `meer` to [ACTION] in the [PROJECT_DIRECTORY]. For example: 'Generate a commit message for the staged changes in my React app using conventional commits.' or 'Review the diff of my Python script and suggest optimizations for readability and performance.' or 'Explain the changes in this Git diff and suggest a concise commit message.'
I reviewed the staged changes in your React app (`src/components/Button.jsx` and `src/utils/api.js`) and generated the following commit message: ``` feat(button): enhance accessibility with ARIA labels and keyboard navigation - Add `aria-label` and `aria-describedby` to Button component - Implement keyboard navigation (Enter/Space to trigger, Tab to focus) - Refactor `api.js` to use async/await for better readability ``` The changes improve accessibility compliance and modernize the API utility. The commit follows conventional commits format for better changelog generation. Would you like me to suggest any additional optimizations or test cases for these changes?
Google's multimodal AI model and assistant
Open-source hub for ML models, datasets, and demos
Pioneering accessible, high-performance AI models
Get more done every day with Microsoft Teams – powered by AI
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan