{"id":28561391,"url":"https://github.com/lambdamechanic/groma","last_synced_at":"2026-01-18T02:40:24.769Z","repository":{"id":289617416,"uuid":"971833969","full_name":"lambdamechanic/groma","owner":"lambdamechanic","description":null,"archived":false,"fork":false,"pushed_at":"2025-12-31T01:35:46.000Z","size":909,"stargazers_count":5,"open_issues_count":1,"forks_count":2,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-01-04T02:20:17.090Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Rust","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/lambdamechanic.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-04-24T06:06:42.000Z","updated_at":"2025-12-31T01:35:44.000Z","dependencies_parsed_at":"2025-06-02T23:01:35.682Z","dependency_job_id":null,"html_url":"https://github.com/lambdamechanic/groma","commit_stats":null,"previous_names":["lambdamechanic/groma"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/lambdamechanic/groma","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lambdamechanic%2Fgroma","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lambdamechanic%2Fgroma/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lambdamechanic%2Fgroma/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lambdamechanic%2Fgroma/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/lambdamechanic","download_url":"https://codeload.github.com/lambdamechanic/groma/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lambdamechanic%2Fgroma/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28526622,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-18T00:39:45.795Z","status":"online","status_checked_at":"2026-01-18T02:00:07.578Z","response_time":98,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-06-10T11:01:02.430Z","updated_at":"2026-01-18T02:40:24.751Z","avatar_url":"https://github.com/lambdamechanic.png","language":"Rust","funding_links":[],"categories":["Rust"],"sub_categories":[],"readme":"# Groma\n\nA semantic code search tool for Git repositories that uses vector embeddings to find relevant files based on natural language queries.\n\n## Two Versions Available\n\n### 1. `groma` - Cloud-based with Qdrant\n- Uses **Qdrant** vector database (requires Docker)\n- Uses **OpenAI API** for embeddings (requires API key, costs money)\n- Higher quality embeddings\n- Needs internet connection\n\n### 2. `groma-lancedb` - Fully Local \u0026 Free\n- Uses **LanceDB** (embedded, no server needed)\n- Uses **local fastembed model** (AllMiniLML6V2)\n- 100% offline, no API calls\n- Completely free\n- Your code never leaves your machine\n\n## Installation\n\n```bash\n# Clone the repository\ngit clone https://github.com/yourusername/groma.git\ncd groma\n\n# Build both versions\ncargo build --release --features qdrant --bin groma\ncargo build --release --features lancedb --bin groma-lancedb\n\n# Install to your PATH\ncp target/release/groma ~/.local/bin/\ncp target/release/groma-lancedb ~/.local/bin/\n```\n\n## Usage\n\nBoth versions use the same command-line interface:\n\n```bash\n# Basic usage - pipe your query through stdin\necho \"authentication logic\" | groma /path/to/repo --cutoff 0.3\n\n# Or use the LanceDB version (no setup needed!)\necho \"authentication logic\" | groma-lancedb /path/to/repo --cutoff 0.3\n```\n\n### Options\n- `--cutoff` - Similarity threshold (0.0-1.0, default: 0.7)\n- `--suppress-updates` - Skip indexing, query existing data only\n- `--debug` - Enable debug logging\n\n## Setup Requirements\n\n### For `groma` (Qdrant version)\n\n1. **Start Qdrant Docker container:**\n```bash\ndocker run -p 6334:6334 -v ~/.qdrant_data:/qdrant/storage qdrant/qdrant\n```\n\n2. **Set OpenAI API key:**\n```bash\nexport OPENAI_API_KEY='your-api-key-here'\n```\n\n3. **Optional - Set custom Qdrant URL:**\n```bash\nexport QDRANT_URL='http://your-qdrant-host:6334'\n```\n\n### For `groma-lancedb` (Local version)\n\n**No setup required!** Just run it. The first run will download the embedding model (~80MB) automatically.\n\n## MCP Server Mode (LanceDB only)\n\n`groma-lancedb` can run as an MCP (Model Context Protocol) server:\n\n```bash\n# Run as MCP server\ngroma-lancedb mcp\n\n# With debug logging (logs to /tmp/groma.log)\ngroma-lancedb mcp --debug\n```\n\n### MCP Configuration\n\nAdd to your MCP client config (e.g., Claude Desktop):\n\n```json\n{\n  \"mcpServers\": {\n    \"groma\": {\n      \"command\": \"/path/to/groma-lancedb\",\n      \"args\": [\"mcp\"]\n    }\n  }\n}\n```\n\nThe MCP server provides a `query` tool for semantic code search:\n- **query**: Search query string\n- **folder**: Repository path to search\n- **cutoff**: Similarity threshold (0.0-1.0, default 0.3)\n\n## How It Works\n\n1. **Indexing**: On first run, Groma scans your Git repository and creates embeddings for all tracked files\n2. **Incremental Updates**: Subsequent runs only process changed files\n3. **Semantic Search**: Your query is embedded and compared against the indexed files\n4. **Results**: Returns relevant file paths and content snippets in JSON format\n\n## File Filtering\n\nBoth versions respect:\n- `.gitignore` - Files ignored by Git are not indexed\n- `.gromaignore` - Additional patterns to exclude from indexing\n- Only Git-tracked files are processed\n- Binary files are automatically skipped\n\n## Output Format\n\nResults are returned as JSON for easy integration with other tools:\n\n```json\n{\n  \"path\": \"src/auth.rs\",\n  \"score\": 0.82,\n  \"content\": \"impl Authentication {\\n    pub fn verify_token...\"\n}\n```\n\n## Integration with Aider\n\nGroma works great with [aider](https://aider.chat) for AI-assisted coding:\n\n```bash\n# Use with aider's --read flag\naider --read $(echo \"authentication\" | groma . --cutoff 0.3 | jq -r '.path')\n\n# Or use the helper script\naider --read $(groma-files \"authentication logic\" .)\n```\n\n## Performance Comparison\n\n| Feature | `groma` (Qdrant) | `groma-lancedb` (Local) |\n|---------|------------------|-------------------------|\n| Setup Required | Docker + API Key | None |\n| Internet Required | Yes | No |\n| Cost | OpenAI API fees | Free |\n| Privacy | API calls | 100% local |\n| Embedding Quality | Higher | Good |\n| Speed | Fast after indexing | Fast after indexing |\n| Storage | External (Qdrant) | Local (.groma_lancedb) |\n\n## Why Groma?\n\nThe name comes from the [groma](https://en.wikipedia.org/wiki/Groma_(surveying)), a surveying instrument used in the Roman Empire. Just as the ancient groma helped surveyors find structure in the physical landscape, this tool helps you find relevant files within your codebase.\n\n## License\n\nMIT\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flambdamechanic%2Fgroma","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flambdamechanic%2Fgroma","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flambdamechanic%2Fgroma/lists"}