Kodit

A code and document intelligence server that indexes Git repositories and provides search through MCP and REST APIs.

AI coding assistants work better when they have access to real examples from your codebase. Kodit indexes your repositories, splits source files into searchable snippets, and serves them to any MCP-compatible assistant. When your assistant needs to write new code, it queries Kodit first and gets back relevant, up-to-date examples drawn from your own projects.

Kodit also handles documents. PDFs, Word files, PowerPoint decks, and spreadsheets are rasterized and indexed so you can search across both code and documentation in one place.

What you get:

• Multiple search strategies including BM25 keyword search, semantic vector search, regex grep, and visual document search, each exposed as a separate MCP tool so your assistant picks the right approach for each query
• MCP server that works with Claude Code, Cursor, Cline, Kilo Code, and any other MCP-compatible assistant
• REST API for programmatic access to search, repositories, enrichments, and indexing status
• AI enrichments (optional) including architecture docs, API docs, database schema detection, cookbook examples, and commit summaries, all generated by an LLM
• Document intelligence with visual search across PDF pages, Office documents, and images using multimodal embeddings
• No external dependencies required for basic operation, with a built-in embedding model and SQLite storage

Quickstart

Docker (recommended)

docker run -p 8080:8080 registry.helix.ml/helix/kodit:latest

This starts Kodit with SQLite storage and a built-in embedding model. No API keys needed.

Pre-built binaries

Download a binary from the releases page, then:

chmod +x kodit
./kodit serve

Verify it works

Open the interactive API docs at http://localhost:8080/docs.

Or index a small repository and run a search:

# Index a repository
curl http://localhost:8080/api/v1/repositories \
  -X POST -H "Content-Type: application/json" \
  -d '{
    "data": {
      "type": "repository",
      "attributes": {
        "remote_uri": "https://gist.github.com/philwinder/7aa38185e20433c04c533f2b28f4e217.git"
      }
    }
  }'

# Check indexing progress
curl http://localhost:8080/api/v1/repositories/1/status

# Search (once indexing is complete)
curl http://localhost:8080/api/v1/search \
  -X POST -H "Content-Type: application/json" \
  -d '{
    "data": {
      "type": "search",
      "attributes": {
        "keywords": ["orders"],
        "text": "code to get all orders"
      }
    }
  }'

Connecting to AI Assistants

Kodit exposes an MCP endpoint at /mcp. Connect your assistant to start using Kodit as a code search tool.

Claude Code

claude mcp add --transport http kodit http://localhost:8080/mcp

Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "kodit": {
      "url": "http://localhost:8080/mcp"
    }
  }
}

Cline

Add to the MCP Servers configuration (Remote Servers tab):

{
  "mcpServers": {
    "kodit": {
      "autoApprove": [],
      "disabled": false,
      "timeout": 60,
      "type": "streamableHttp",
      "url": "http://localhost:8080/mcp"
    }
  }
}

Kilo Code

Add to the MCP configuration (Edit Project/Global MCP):

{
  "mcpServers": {
    "kodit": {
      "type": "streamable-http",
      "url": "http://localhost:8080/mcp",
      "alwaysAllow": [],
      "disabled": false
    }
  }
}

Replace http://localhost:8080 with your server URL if running remotely.

Encouraging assistants to use Kodit

Some assistants may not call Kodit tools automatically. Add this to your project rules or system prompt to enforce usage:

For every request that involves writing or modifying code, the assistant's first
action must be to call the kodit search MCP tools. Only produce or edit code after
the tool call returns results.

In Cursor, save this as .cursor/rules/kodit.mdc with alwaysApply: true frontmatter.

MCP Tools

Kodit exposes these tools to connected AI assistants:

Tool	Description
kodit_repositories	List all indexed repositories
kodit_semantic_search	Semantic similarity search across code
kodit_keyword_search	BM25 keyword search
kodit_visual_search	Search document page images
kodit_grep	Regex pattern matching
kodit_ls	List files by glob pattern
kodit_read_resource	Read file content by URI
kodit_architecture_docs	Architecture documentation for a repo
kodit_api_docs	Public API documentation
kodit_database_schema	Database schema documentation
kodit_cookbook	Usage examples and patterns
kodit_commit_description	Commit description
kodit_wiki	Wiki table of contents
kodit_wiki_page	Read a specific wiki page
kodit_version	Server version

The enrichment tools (architecture_docs, api_docs, database_schema, cookbook, wiki, commit_description) require an LLM provider to be configured. See Enrichment Providers under Configuration Reference.

Go Library

Kodit can be embedded directly as a Go library. This is how Helix integrates Kodit into its platform.

import "github.com/helixml/kodit"

client, err := kodit.New(
    kodit.WithSQLite(".kodit/data.db"),
)
if err != nil {
    log.Fatal(err)
}
defer client.Close()

// Index a repository
_, _, err = client.Repositories.Add(ctx, &service.RepositoryAddParams{
    URL: "https://github.com/kubernetes/kubernetes",
})

// Search
results, err := client.Search.Query(ctx, "create a deployment",
    service.WithLimit(10),
)

for _, result := range results.Enrichments() {
    fmt.Println(result.Subtype(), result.Content())
}

Library options

Option	Description
WithSQLite(path)	Use SQLite for storage
WithPostgresVectorchord(dsn)	Use PostgreSQL with VectorChord
WithOpenAI(apiKey)	OpenAI for embeddings and text
WithAnthropic(apiKey)	Anthropic Claude for text (needs separate embedding provider)
WithTextProvider(p)	Custom text generation provider
WithEmbeddingProvider(p)	Custom embedding provider
WithRAGPipeline()	Skip LLM enrichments, index and search only
WithFullPipeline()	Require all enrichments (errors without a text provider)
WithDataDir(dir)	Data directory (default: ~/.kodit)
WithCloneDir(dir)	Repository clone directory
WithAPIKeys(keys...)	API keys for HTTP authentication
WithWorkerCount(n)	Number of background workers (default: 1)
WithPeriodicSyncConfig(cfg)	Automatic repository sync settings

Search options

Option	Description
WithSemanticWeight(w)	Weight for semantic vs keyword search (0.0 to 1.0)
WithLimit(n)	Maximum number of results
WithOffset(n)	Offset for pagination
WithLanguages(langs...)	Filter by programming languages
WithRepositories(ids...)	Filter by repository IDs
WithMinScore(score)	Minimum score threshold
WithEnrichmentTypes(types...)	Filter results to specific enrichment types
WithSnippets(include)	Include code snippets in results
WithDocuments(include)	Include enrichment documents in results

Go HTTP client

A generated HTTP client is available for calling a remote Kodit server from Go:

go get github.com/helixml/kodit/clients/go

import koditclient "github.com/helixml/kodit/clients/go"

client, err := koditclient.NewClient("https://kodit.example.com")

// List repositories
resp, err := client.GetRepositories(ctx, nil)

// Search
text := "create a deployment"
resp, err := client.PostSearch(ctx, koditclient.PostSearchJSONRequestBody{
    Data: &koditclient.DtoSearchData{
        Attributes: &koditclient.DtoSearchAttributes{
            Text: &text,
        },
    },
})

Types are auto-generated from the OpenAPI spec. See the interactive API docs at /docs for the full endpoint list.

Production Deployment

For production use, deploy with PostgreSQL (VectorChord) for scalable vector search and a dedicated LLM provider for enrichments.

Docker Compose

Save this as docker-compose.yaml:

services:
  kodit:
    image: registry.helix.ml/helix/kodit:latest
    ports:
      - "8080:8080"
    command: ["serve"]
    restart: unless-stopped
    depends_on:
      - vectorchord
    environment:
      DATA_DIR: /data
      DB_URL: postgresql://postgres:mysecretpassword@vectorchord:5432/kodit

      # Enrichment LLM (optional, enables AI-generated docs)
      ENRICHMENT_ENDPOINT_BASE_URL: http://ollama:11434
      ENRICHMENT_ENDPOINT_MODEL: ollama/qwen3:1.7b

      # External embedding provider (optional, replaces built-in model)
      # EMBEDDING_ENDPOINT_API_KEY: sk-proj-xxxx
      # EMBEDDING_ENDPOINT_MODEL: openai/text-embedding-3-small

      LOG_LEVEL: INFO
      API_KEYS: ${KODIT_API_KEYS:-}
    volumes:
      - kodit-data:/data

  vectorchord:
    image: tensorchord/vchord-suite:pg17-20250601
    environment:
      POSTGRES_DB: kodit
      POSTGRES_PASSWORD: mysecretpassword
    volumes:
      - vectorchord-data:/var/lib/postgresql/data
    restart: unless-stopped

volumes:
  kodit-data:
  vectorchord-data:

Kubernetes

apiVersion: apps/v1
kind: Deployment
metadata:
  name: vectorchord
spec:
  replicas: 1
  selector:
    matchLabels:
      app: vectorchord
  template:
    metadata:
      labels:
        app: vectorchord
    spec:
      containers:
        - name: vectorchord
          image: tensorchord/vchord-suite:pg17-20250601
          env:
            - name: POSTGRES_DB
              value: kodit
            - name: POSTGRES_PASSWORD
              value: mysecretpassword
          ports:
            - containerPort: 5432
---
apiVersion: v1
kind: Service
metadata:
  name: vectorchord
spec:
  selector:
    app: vectorchord
  ports:
    - port: 5432
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: kodit
spec:
  replicas: 1
  selector:
    matchLabels:
      app: kodit
  template:
    metadata:
      labels:
        app: kodit
    spec:
      containers:
        - name: kodit
          image: registry.helix.ml/helix/kodit:latest # pin to a specific version
          args: ["serve"]
          env: [] # see Configuration Reference for environment variables
          ports:
            - containerPort: 8080
          readinessProbe:
            httpGet:
              path: /
              port: 8080
            initialDelaySeconds: 10
            periodSeconds: 5
---
apiVersion: v1
kind: Service
metadata:
  name: kodit
spec:
  type: LoadBalancer
  selector:
    app: kodit
  ports:
    - port: 8080

Authentication

Set the API_KEYS environment variable to a comma-separated list of keys. Write endpoints (creating repositories, triggering syncs) require a valid key in the Authorization: Bearer <key> header. Search endpoints are open by default.

Configuration Reference

Configuration is done through environment variables. You can also use a .env file:

kodit serve --env-file .env

Server

Variable	Default	Description
HOST	0.0.0.0	Listen address
PORT	8080	Listen port
DATA_DIR	~/.kodit	Data directory for models, clones, and database
DB_URL	(empty)	PostgreSQL connection string (uses SQLite if empty)
LOG_LEVEL	INFO	Logging verbosity: DEBUG, INFO, WARN, ERROR
LOG_FORMAT	pretty	Log format: pretty or json
API_KEYS	(empty)	Comma-separated API keys for write endpoints
WORKER_COUNT	1	Number of background workers
SEARCH_LIMIT	10	Default search result limit
DISABLE_TELEMETRY	false	Disable anonymous usage telemetry
HTTP_CACHE_DIR	(empty)	Directory for caching HTTP POST responses to disk; avoids repeated API calls during development
REPORTING_LOG_TIME_INTERVAL	5	Progress reporting interval in seconds

Embedding Provider

These configure an external embedding model. If unset, Kodit uses its built-in model.

Variable	Default	Description
EMBEDDING_ENDPOINT_BASE_URL	(empty)	Base URL of embedding service
EMBEDDING_ENDPOINT_MODEL	(empty)	Model identifier
EMBEDDING_ENDPOINT_API_KEY	(empty)	API key
EMBEDDING_ENDPOINT_MAX_TOKENS	0	Max tokens per request (0 = provider default)
EMBEDDING_ENDPOINT_MAX_BATCH_CHARS	16000	Max total characters per embedding batch
EMBEDDING_ENDPOINT_MAX_BATCH_SIZE	1	Max items per batch
EMBEDDING_ENDPOINT_TIMEOUT	60	Request timeout in seconds
EMBEDDING_ENDPOINT_NUM_PARALLEL_TASKS	1	Concurrent embedding requests
EMBEDDING_ENDPOINT_EXTRA_PARAMS	(empty)	JSON-encoded extra parameters for the embedding provider
EMBEDDING_ENDPOINT_QUERY_INSTRUCTION	(empty)	Instruction prepended to queries for asymmetric retrieval
EMBEDDING_ENDPOINT_DOCUMENT_INSTRUCTION	(empty)	Instruction prepended to documents for asymmetric retrieval
EMBEDDING_ENDPOINT_SOCKET_PATH	(empty)	Unix socket path for local provider (alternative to BASE_URL)
EMBEDDING_ENDPOINT_MAX_RETRIES	5	Maximum retry attempts on request failure
EMBEDDING_ENDPOINT_INITIAL_DELAY	2.0	Initial retry delay in seconds
EMBEDDING_ENDPOINT_BACKOFF_FACTOR	2.0	Retry backoff multiplier

Vision Embedding Provider

These configure a remote service for image and text vision embeddings. If unset, Kodit uses its built-in SigLIP2 model.

Variable	Default	Description
VISION_EMBEDDING_ENDPOINT_BASE_URL	(empty)	Base URL of vision embedding service
VISION_EMBEDDING_ENDPOINT_MODEL	(empty)	Model identifier
VISION_EMBEDDING_ENDPOINT_API_KEY	(empty)	API key
VISION_EMBEDDING_ENDPOINT_MAX_TOKENS	0	Max tokens per request (0 = provider default)
VISION_EMBEDDING_ENDPOINT_MAX_BATCH_CHARS	16000	Max total characters per embedding batch
VISION_EMBEDDING_ENDPOINT_MAX_BATCH_SIZE	1	Max items per batch
VISION_EMBEDDING_ENDPOINT_TIMEOUT	60	Request timeout in seconds
VISION_EMBEDDING_ENDPOINT_NUM_PARALLEL_TASKS	1	Concurrent vision embedding requests
VISION_EMBEDDING_ENDPOINT_EXTRA_PARAMS	(empty)	JSON-encoded extra parameters for the vision embedding provider
VISION_EMBEDDING_ENDPOINT_QUERY_INSTRUCTION	(empty)	Instruction prepended to queries for asymmetric retrieval
VISION_EMBEDDING_ENDPOINT_DOCUMENT_INSTRUCTION	(empty)	Instruction prepended to documents for asymmetric retrieval
VISION_EMBEDDING_ENDPOINT_SOCKET_PATH	(empty)	Unix socket path for local provider (alternative to BASE_URL)
VISION_EMBEDDING_ENDPOINT_MAX_RETRIES	5	Maximum retry attempts on request failure
VISION_EMBEDDING_ENDPOINT_INITIAL_DELAY	2.0	Initial retry delay in seconds
VISION_EMBEDDING_ENDPOINT_BACKOFF_FACTOR	2.0	Retry backoff multiplier

Enrichment Providers

These configure an LLM for generating architecture docs, API docs, database schemas, cookbooks, commit summaries, and wiki pages. Without this, Kodit indexes and searches code but does not generate any AI documentation.

Variable	Default	Description
ENRICHMENT_ENDPOINT_BASE_URL	(empty)	Base URL of LLM service
ENRICHMENT_ENDPOINT_MODEL	(empty)	Model identifier
ENRICHMENT_ENDPOINT_API_KEY	(empty)	API key
ENRICHMENT_ENDPOINT_NUM_PARALLEL_TASKS	1	Concurrent enrichment requests
ENRICHMENT_ENDPOINT_TIMEOUT	60	Request timeout in seconds
ENRICHMENT_ENDPOINT_EXTRA_PARAMS	(empty)	JSON-encoded extra parameters for the LLM
ENRICHMENT_ENDPOINT_MAX_TOKENS	0	Max tokens per response (0 = provider default)
ENRICHMENT_ENDPOINT_SOCKET_PATH	(empty)	Unix socket path for local provider (alternative to BASE_URL)
ENRICHMENT_ENDPOINT_MAX_RETRIES	5	Maximum retry attempts on request failure
ENRICHMENT_ENDPOINT_INITIAL_DELAY	2.0	Initial retry delay in seconds
ENRICHMENT_ENDPOINT_BACKOFF_FACTOR	2.0	Retry backoff multiplier
ENRICHMENT_ENDPOINT_MAX_BATCH_CHARS	16000	Max total characters per batch
ENRICHMENT_ENDPOINT_MAX_BATCH_SIZE	1	Max items per batch
ENRICHMENT_ENDPOINT_QUERY_INSTRUCTION	(empty)	Instruction prepended to queries for asymmetric retrieval
ENRICHMENT_ENDPOINT_DOCUMENT_INSTRUCTION	(empty)	Instruction prepended to documents for asymmetric retrieval

Enrichment is typically the slowest part of indexing because each enrichment requires a round-trip to the LLM provider. Increase NUM_PARALLEL_TASKS to speed things up, but respect your provider's rate limits. Start low and increase over time.

Provider examples:

# OpenAI
ENRICHMENT_ENDPOINT_BASE_URL=https://api.openai.com/v1
ENRICHMENT_ENDPOINT_MODEL=gpt-4o-mini
ENRICHMENT_ENDPOINT_API_KEY=sk-proj-xxxx

# Ollama (local)
ENRICHMENT_ENDPOINT_BASE_URL=http://localhost:11434
ENRICHMENT_ENDPOINT_MODEL=ollama/qwen3:1.7b

# Helix (private cloud)
ENRICHMENT_ENDPOINT_BASE_URL=https://app.helix.ml/v1
ENRICHMENT_ENDPOINT_MODEL=Qwen/Qwen3-8B
ENRICHMENT_ENDPOINT_API_KEY=your-helix-key

Periodic Sync

Variable	Default	Description
PERIODIC_SYNC_ENABLED	true	Auto-sync repositories on an interval
PERIODIC_SYNC_INTERVAL_SECONDS	1800	Sync interval (default: 30 minutes)
PERIODIC_SYNC_RETRY_ATTEMPTS	3	Retry count on sync failure

Chunking

Variable	Default	Description
CHUNK_SIZE	1500	Characters per chunk
CHUNK_OVERLAP	200	Overlap between adjacent chunks
CHUNK_MIN_SIZE	50	Minimum chunk size

REST API

The full API is documented interactively at /docs on a running Kodit instance. The OpenAPI 3.0 specification is available at /docs/openapi.json.

Key endpoints:

Method	Path	Description
POST	/api/v1/repositories	Add a repository for indexing
GET	/api/v1/repositories	List indexed repositories
GET	/api/v1/repositories/{id}/status	Indexing progress
POST	/api/v1/repositories/{id}/sync	Trigger a sync
DELETE	/api/v1/repositories/{id}	Remove a repository
POST	/api/v1/search	Combined search (keyword + semantic)
GET	/api/v1/search/semantic	Semantic search only
GET	/api/v1/search/keyword	Keyword search only
GET	/api/v1/search/visual	Visual search on document pages
GET	/api/v1/search/grep	Regex pattern search
GET	/api/v1/search/ls	List files by glob

All write endpoints require an Authorization: Bearer <key> header when API_KEYS is set.

How Indexing Works

When you add a repository, Kodit runs a pipeline:

1. Clone the Git repository to local storage
2. Scan commits, branches, and tags to extract metadata
3. Extract snippets by splitting source files into overlapping text chunks
4. Build search indexes with BM25 (keyword) and vector embeddings (semantic)
5. Generate enrichments (if an LLM provider is configured): architecture docs, API docs, database schemas, cookbook examples, commit summaries, and wiki pages

Kodit tracks which files have changed between syncs and only reprocesses modified content. Repositories sync automatically on a configurable interval (default: every 30 minutes).

Supported sources

Kodit indexes any Git repository accessible via HTTPS, SSH, or the Git protocol. This includes GitHub, GitLab, Bitbucket, Azure DevOps, and self-hosted servers.

Private repositories

Private repositories are supported through personal access tokens or SSH keys:

# HTTPS with token
https://username:token@github.com/username/repo.git

# SSH (ensure your SSH key is configured)
git@github.com:username/repo.git

Privacy

Kodit respects .gitignore and .noindex files. Files matching these patterns are excluded from indexing.

Storage Backends

SQLite (default)

No configuration needed. Kodit creates a SQLite database in the data directory with FTS5 for keyword search and in-process vector storage. Good for single-user and small-team deployments.

PostgreSQL with VectorChord

For larger deployments, use PostgreSQL with the VectorChord extension. This provides scalable vector search and concurrent access. Set the DB_URL environment variable to your connection string.

The recommended Docker image is tensorchord/vchord-suite:pg17-20250601, which bundles PostgreSQL 17 with VectorChord, vchord_bm25, and pg_tokenizer.

Building from Source

git clone https://github.com/helixml/kodit.git
cd kodit
make tools          # Install development tools
make download-model # Download the built-in embedding model
make build          # Build the binary
./bin/kodit version
./bin/kodit serve

Run the tests:

make test                         # All tests
make test PKG=./internal/foo/...  # Specific package
make check                        # Format, vet, lint, and test

Troubleshooting

MCP connection error after restart: If you see No valid session ID provided after restarting the Kodit server, reload the MCP client in your assistant. MCP sessions do not survive server restarts.

No search results: Check that indexing has completed by calling GET /api/v1/repositories/{id}/status. If status shows errors, check the server logs with LOG_LEVEL=DEBUG.

Enrichments not generating: Enrichments require an LLM provider. Check that ENRICHMENT_ENDPOINT_BASE_URL and ENRICHMENT_ENDPOINT_MODEL are set. Without these, Kodit indexes and searches code but does not generate AI documentation.

Telemetry

Kodit collects limited anonymous telemetry (usage metadata only, no user data) to guide development. Disable it with:

DISABLE_TELEMETRY=true

Commercial Support

Helix provides a managed platform built on Kodit with additional features including a management UI, repository browsing, team collaboration, and hosted infrastructure. For commercial support or enterprise integration, contact founders@helix.ml.

Contributing

See CONTRIBUTING.md for guidelines.

License

Apache 2.0