{"id":"c21c19af-686d-4e92-88fd-5aa368885d4b","slug":"cie","kind":"mcp","name":"cie","summary":"Code Intelligence Engine — indexes your codebase and gives AI assistants deep understanding via MCP (semantic search, call graphs, 20+ tools)","owner":"kraklabs","repoUrl":"https://github.com/kraklabs/cie","license":"AGPL-3.0","licenseTier":"strong-copyleft","trust":"community","stars":19,"match":null,"fewStars":false,"format":"mcp","createdAt":null,"addedAt":"2026-10-14T05:18:28.284Z","logoUrl":"https://avatars.githubusercontent.com/u/238871896?v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["developer-tools","search","memory-knowledge"],"tags":["call-graph","claude-code","code-analysis","code-intelligence","cursor","developers-tools","semantic-search","tree-sitter"],"runtime":"go","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-14T09:06:07.900Z","sample":false,"url":"https://conduits.dev/tools/cie","language":"Go","transports":["stdio"],"install":{},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"<div align=\"center\">\n  <h1>CIE - Code Intelligence Engine</h1>\n  <p><strong>20+ MCP tools that give AI agents semantic code search, call graph analysis, and endpoint discovery — 100% local, indexes 100k LOC in seconds.</strong></p>\n\n  <img src=\"docs/cie-demo.gif\" alt=\"CIE Demo\" width=\"800\">\n\n  <blockquote>\n    <p><em>In practice: Tracing a call graph through MIE's codebase, Claude Code needed 34 tool calls without CIE. With CIE, it needed 3. Results vary by query, but structured code intelligence consistently reduces agent round-trips.</em></p>\n  </blockquote>\n\n  [![Release](https://github.com/kraklabs/cie/actions/workflows/release.yml/badge.svg)](https://github.com/kraklabs/cie/actions/workflows/release.yml)\n  [![codecov](https://codecov.io/gh/kraklabs/cie/branch/main/graph/badge.svg)](https://codecov.io/gh/kraklabs/cie)\n  [![Go Report Card](https://goreportcard.com/badge/github.com/kraklabs/cie)](https://goreportcard.com/report/github.com/kraklabs/cie)\n  [![Go Version](https://img.shields.io/github/go-mod/go-version/kraklabs/cie)](go.mod)\n  [![License](https://img.shields.io/badge/license-AGPL%20v3-blue.svg)](LICENSE)\n\n  <p>\n    <a href=\"#quick-start\">Quick Start</a> •\n    <a href=\"#features\">Features</a> •\n    <a href=\"#documentation\">Documentation</a> •\n    <a href=\"#support\">Support</a>\n  </p>\n</div>\n\n---\n\nCIE indexes your codebase and provides semantic search, call graph analysis, and AI-powered code understanding through the Model Context Protocol (MCP).\n\n## Why CIE?\n\n- **Semantic Search** - Find code by meaning, not just text matching\n- **Call Graph Analysis** - Trace execution paths including interface dispatch resolution\n- **MCP Native** - Works seamlessly with Claude Code, Cursor, and any MCP client\n- **Fast** - Indexes 100k LOC in seconds, queries in milliseconds\n- **Private** - All data stays local, your code never leaves your machine\n- **Accurate** - Keyword boosting ensures relevant results for function searches\n\n## Installation\n\n| Method | Command |\n|--------|---------|\n| **Homebrew** | `brew tap kraklabs/cie && brew install cie` |\n| **Install Script** | `curl -sSL https://raw.githubusercontent.com/kraklabs/cie/main/install.sh \\| sh` |\n| **GitHub Releases** | [Download binary](https://github.com/kraklabs/cie/releases/latest) |\n\n## Features\n\n### Semantic Code Search\n\nFind code by meaning, not keywords:\n\n```bash\n# Ask: \"Where is authentication middleware?\"\n# Use cie_semantic_search tool via MCP\n```\n\n**Example output:**\n```\n[95%] AuthMiddleware (internal/http/auth.go:42)\n[76%] ValidateToken (internal/auth/jwt.go:103)\n```\n\n### Call Graph Analysis\n\nTrace how execution reaches any function:\n\n```bash\n# Question: \"How does main() reach database.Connect()?\"\n# Use cie_trace_path tool\n```\n\n**Example output:**\n```\nmain → InitApp → SetupDatabase → database.Connect\n  ├─ File: cmd/server/main.go:25\n  ├─ File: internal/app/init.go:42\n  └─ File: internal/database/setup.go:18\n```\n\n### HTTP Endpoint Discovery\n\nList all API endpoints automatically:\n\n```bash\n# Use cie_list_endpoints tool\n```\n\n**Example output:**\n```\n[GET]    /api/v1/users          → HandleGetUsers\n[POST]   /api/v1/users          → HandleCreateUser\n[DELETE] /api/v1/users/:id      → HandleDeleteUser\n```\n\n### Multi-Language Support\n\nSupports Go, Python, JavaScript, TypeScript, and more through Tree-sitter parsers.\n\n## Quick Start\n\n### 1. Install the CLI\n\n**Homebrew (macOS/Linux):**\n```bash\nbrew tap kraklabs/cie\nbrew install cie\n```\n\n**Script:**\n```bash\ncurl -sSL https://raw.githubusercontent.com/kraklabs/cie/main/install.sh | sh\n```\n\n**Manual download:**\nDownload from [GitHub Releases](https://github.com/kraklabs/cie/releases/latest)\n\n### 2. Index Your Repository\n\n```bash\ncd /path/to/your/repo\ncie init -y    # Initialize project configuration\ncie index      # Index the codebase (works without Ollama too)\n```\n\n**Example output:**\n```\nProject: your-repo-name\nFiles: 1,234\nFunctions: 5,678\nTypes: 890\nLast indexed: 2 minutes ago\n```\n\n> **Note:** CIE works without Ollama -- you'll have access to 20+ tools including grep, call graph, function finder, and more. Semantic search requires embeddings from Ollama or another provider.\n\n### Management Commands\n\n| Command | Description |\n|---------|-------------|\n| `cie init -y` | Initialize project configuration |\n| `cie index` | Index (or re-index) the codebase |\n| `cie reset --yes` | Delete all indexed data for the project |\n\n### MCP Server Mode\n\nCIE can run as an MCP server for integration with Claude Code:\n\n```bash\ncie --mcp\n```\n\nConfigure in your Claude Code settings:\n\n```json\n{\n  \"mcpServers\": {\n    \"cie\": {\n      \"command\": \"cie\",\n      \"args\": [\"--mcp\"]\n    }\n  }\n}\n```\n\n## Configuration\n\nCIE uses a YAML configuration file (`.cie/project.yaml`):\n\n```yaml\nversion: \"1\"\nproject_id: my-project\nembedding:\n  provider: ollama\n  base_url: http://localhost:11434\n  model: nomic-embed-text\n```\n\n> **Embeddings are optional.** CIE works without Ollama or any embedding provider. You get full access to all structural tools (grep, call graph, function finder, etc.). Only semantic search (`cie_semantic_search`) requires embeddings.\n\nYou can also configure an LLM for `cie_analyze` narrative generation:\n\n```yaml\n# Optional: LLM for cie_analyze narrative generation\nllm:\n  enabled: true\n  base_url: http://localhost:11434  # Ollama\n  model: llama3\n  # For OpenAI: base_url: https://api.openai.com/v1, model: gpt-4o-mini\n```\n\n**Note:** The `llm` section is optional. Without it, `cie_analyze` returns raw analysis data. With it configured, you get synthesized narrative summaries.\n\n## MCP Tools\n\nWhen running as an MCP server, CIE provides 20+ tools organized by category:\n\n### Navigation & Search\n\n| Tool | Description |\n|------|-------------|\n| `cie_grep` | Fast literal text search (no regex) |\n| `cie_semantic_search` | Meaning-based search using embeddings |\n| `cie_find_function` | Find functions by name (handles receiver syntax) |\n| `cie_find_type` | Find types/interfaces/structs |\n| `cie_find_similar_functions` | Find functions with similar names |\n| `cie_list_files` | List indexed files with filters |\n| `cie_list_functions_in_file` | List all functions in a file |\n\n### Call Graph Analysis\n\n| Tool | Description |\n|------|-------------|\n| `cie_find_callers` | Find what calls a function |\n| `cie_find_callees` | Find what a function calls |\n| `cie_trace_path` | Trace call paths from entry points to target |\n| `cie_get_call_graph` | Get complete call graph for a function |\n\n### Code Understanding\n\n| Tool | Description |\n|------|-------------|\n| `cie_analyze` | Architectural analysis (LLM narrative optional) |\n| `cie_get_function_code` | Get function source code |\n| `cie_directory_summary` | Get directory overview with main functions |\n| `cie_find_implementations` | Find types that implement an interface |\n| `cie_get_file_summary` | Get summary of all entities in a file |\n\n### HTTP/API Discovery\n\n| Tool | Description |\n|------|-------------|\n| `cie_list_endpoints` | List HTTP/REST endpoints from common Go frameworks |\n| `cie_list_services` | List gRPC services and RPC methods from .proto files |\n\n### Security & Verification\n\n| Tool | Description |\n|------|-------------|\n| `cie_verify_absence` | Verify patterns don't exist (security audits) |\n\n### System\n\n| Tool | Description |\n|------|-------------|\n| `cie_index_status` | Check indexing health and statistics |\n| `cie_search_text` | Regex-based text search in function code |\n| `cie_raw_query` | Execute raw CozoScript queries |\n\n> **For detailed documentation of each tool with examples, see [Tools Reference](docs/tools-reference.md)**\n\n## Data Storage\n\nCIE stores indexed data locally in `<local_data_dir>/<project_id>/` (default: `~/.cie/data/<project_id>/`) using embedded CozoDB with RocksDB backend. This ensures:\n\n- Your code never leaves your machine\n- Fast local queries\n- Persistent index across sessions\n\n## Embedding Providers\n\nCIE supports multiple embedding providers:\n\n| Provider | Configuration |\n|----------|--------------|\n| **Ollama** | `OLLAMA_HOST`, `OLLAMA_EMBED_MODEL` |\n| **OpenAI** | `OPENAI_API_KEY`, `OPENAI_EMBED_MODEL` |\n| **Nomic** | `NOMIC_API_KEY` |\n\n## Documentation\n\n| Guide | Description |\n|-------|-------------|\n| [Getting Started](docs/getting-started.md) | Step-by-step tutorial from installation to first query |\n| [Configuration](docs/configuration.md) | Complete configuration reference |\n| [Tools Reference](docs/tools-reference.md) | All 20+ MCP tools with examples |\n| [Architecture](docs/architecture.md) | How CIE works internally |\n| [MCP Integration](docs/mcp-integration.md) | Setting up with Claude Code, Cursor |\n| [Migration Guide](docs/migration-guide.md) | Migrating from Docker to embedded mode |\n| [Testing Guide](docs/testing.md) | Running tests and adding new tests |\n| [Benchmarks](docs/benchmarks.md) | Performance data and tuning |\n| [Exit Codes](docs/exit-codes.md) | CLI exit codes for scripting |\n| [Troubleshooting](docs/troubleshooting.md) | Common issues and solutions |\n\n## Architecture\n\nCIE uses an embedded architecture -- a single binary handles indexing, querying, and MCP serving with no external services required:\n\n```\n┌──────────────────────────────────────────────────────────────┐\n│  Host Machine                                                 │\n│  ┌──────────────────────────────────────────────────────┐    │\n│  │  CLI `cie`                                            │    │\n│  │  - cie init   → Creates .cie/project.yaml            │    │\n│  │  - cie index  → Parses code, writes to local CozoDB  │    │\n│  │  - cie --mcp  → Reads from local CozoDB              │    │\n│  │                                                       │    │\n│  │  Data: <local_data_dir>/<project>/ (RocksDB)         │    │\n│  └──────────────────────────────────────────────────────┘    │\n│                                                               │\n│  ┌──────────────┐  (optional)                                │\n│  │   Ollama     │  For semantic search embeddings            │\n│  │  :11434      │  Install: brew install ollama              │\n│  └──────────────┘  Model: nomic-embed-text                   │\n└──────────────────────────────────────────────────────────────┘\n```\n\n**Key Components:**\n\n- **CIE CLI**: Single binary handles indexing, querying, and MCP serving\n- **CozoDB + RocksDB**: Embedded database stored locally at `<local_data_dir>/<project>` (default `~/.cie/data/<project>`)\n- **Ollama (optional)**: Local embedding generation for semantic search\n- **Tree-sitter**: Code parsing for Go, Python, JS, TS\n\n**Code Structure:**\n```\ncie/\n├── cmd/cie/           # CLI tool with init, index, query, MCP commands\n├── pkg/\n│   ├── ingestion/     # Tree-sitter parsers and indexing pipeline\n│   ├── tools/         # 20+ MCP tool implementations\n│   ├── llm/           # LLM provider abstractions (OpenAI, Ollama)\n│   ├── cozodb/        # CozoDB wrapper for Datalog queries\n│   └── storage/       # Storage backend interface\n└── docs/              # Documentation\n```\n\nFor in-depth architecture details, see [Architecture Guide](docs/architecture.md).\n\n## Development\n\n### Testing\n\n```bash\n# Run all tests\ngo test ./...\n\n# Run with short flag\ngo test -short ./...\n\n# Run integration tests with CozoDB\ngo test -tags=cozodb ./...\n```\n\nFor detailed testing documentation, see [docs/testing.md](docs/testing.md).\n\n### Writing Tests\n\nUse the CIE testing helpers for easy test setup:\n\n```go\nimport cietest \"github.com/kraklabs/cie/internal/testing\"\n\nfunc TestMyFeature(t *testing.T) {\n    backend := cietest.SetupTestBackend(t)\n    cietest.InsertTestFunction(t, backend, \"func1\", \"MyFunc\", \"file.go\", 10, 20)\n\n    result := cietest.QueryFunctions(t, backend)\n    require.Len(t, result.Rows, 1)\n}\n```\n\n### Building\n\n```bash\n# Build all commands\nmake build-all\n\n# Format code\nmake fmt\n\n# Run linter\nmake lint\n```\n\n## Support\n\nNeed help or want to contribute?\n\n- **Documentation**: [docs/](docs/)\n- **Report Issues**: [GitHub Issues](https://github.com/kraklabs/cie/issues/new?labels=cie)\n- **Discussions**: [GitHub Discussions](https://github.com/kraklabs/cie/discussions)\n- **Email**: support@kraklabs.com\n\n**Before opening an issue:**\n1. Check the [troubleshooting guide](docs/troubleshooting.md)\n2. Search [existing issues](https://github.com/kraklabs/cie/issues?q=label%3Acie)\n3. Include CIE version: `cie --version`\n4. Provide minimal reproduction steps\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.\n\n## CIE Enterprise\n\n**Scale code intelligence across your entire organization.**\n\nCIE Enterprise brings the power of semantic code search and call graph analysis to teams of any size. Built for organizations that demand reliability, security, and collaboration.\n\n### Why Enterprise?\n\n| Feature | Open Source | Enterprise |\n|---------|-------------|------------|\n| Semantic Search | ✅ | ✅ |\n| Call Graph Analysis | ✅ | ✅ |\n| Local Embeddings (768 dim) | ✅ | ✅ |\n| **Distributed Architecture** | — | ✅ |\n| **Team Collaboration** | — | ✅ |\n| **CI/CD Integration** | — | ✅ |\n| **High-Fidelity Embeddings (1536 dim)** | — | ✅ |\n| **Integrated LLMs** | — | ✅ |\n| **Priority Support** | — | ✅ |\n\n### Enterprise Features\n\n**Distributed Architecture**\nDeploy CIE across your infrastructure with a Primary Hub and Edge Caches. All team members connect to the same indexed codebase with millisecond-latency queries worldwide.\n\n**Team Collaboration**\nShare code intelligence across your entire engineering organization. One index, one source of truth—no more siloed knowledge.\n\n**CI/CD Integration**\nAutomatically keep your code index up-to-date with every commit. Native integration with GitHub Actions, GitLab CI, Jenkins, and more.\n\n**High-Fidelity Embeddings**\nOpenAI-powered 1536-dimension embeddings for superior semantic search accuracy. Find exactly what you're looking for, even in massive codebases.\n\n**Integrated LLMs**\nConnect your preferred LLM provider for enhanced code analysis, architectural insights, and natural language queries about your codebase.\n\n**Priority Support**\nDirect access to our engineering team. SLAs, dedicated support channels, and implementation assistance.\n\n### Get Started\n\n**Contact us:** enterprise@kraklabs.com\n\nSchedule a demo to see how CIE Enterprise can transform your team's development workflow.\n\n---\n\n## License\n\nCIE is dual-licensed:\n\n### Open Source License (AGPL v3)\n\nCIE is free and open source under the **GNU Affero General Public License v3.0** (AGPL v3).\n\n**Use CIE for free if:**\n- You're building open source software\n- You can release your modifications under AGPL v3\n- You're okay with the copyleft requirements\n\nSee [LICENSE](LICENSE) for full AGPL v3 terms.\n\n### Commercial License\n\nNeed to use CIE in a closed-source product or service? We offer commercial licenses that remove AGPL requirements.\n\n**Commercial licensing is right for you if:**\n- You want to use CIE in a proprietary product\n- You want to offer CIE as a managed service without releasing your code\n- Your organization's policies prohibit AGPL-licensed software\n- You want to modify CIE without releasing your modifications\n\n**Pricing:** Contact licensing@kraklabs.com for details.\n\nSee [LICENSE.commercial](LICENSE.commercial) for more information.\n\n**Why dual licensing?**\nThis model allows us to:\n- Keep CIE free for the open source community\n- Ensure improvements benefit everyone through AGPL's copyleft\n- Sustainably fund development through commercial licensing\n- Enable enterprise adoption without legal concerns\n\n### Third-Party Components\n\nCIE includes some third-party components with their own licenses:\n- **CozoDB C Headers** (MPL 2.0) - See [THIRD_PARTY_LICENSES.md](THIRD_PARTY_LICENSES.md) for details\n\nThese components are compatible with AGPL v3 and retain their original licenses.\n\n## Related Projects\n\n- [CozoDB](https://github.com/cozodb/cozo) - The embedded database powering CIE\n- [Tree-sitter](https://tree-sitter.github.io/) - Parser generator for code analysis\n- [MCP](https://modelcontextprotocol.io/) - Model Context Protocol specification\n"}