@mhalder/qdrant-mcp-server 1.3.1 → 1.5.0

package/.codecov.yml

@@ -0,0 +1,16 @@
+coverage:
+  status:
+    project:
+      default:
+        target: auto
+        threshold: 5%
+        informational: true
+    patch:
+      default:
+        target: 80%
+        threshold: 5%
+
+comment:
+  layout: "header, diff, flags, components"
+  behavior: default
+  require_changes: false

package/CHANGELOG.md

@@ -1,3 +1,28 @@
+## 1.5.0 (2025-10-30)
+
+* Merge pull request #32 from mhalder/feature/code-vectorization ([50c6cb0](https://github.com/mhalder/qdrant-mcp-server/commit/50c6cb0)), closes [#32](https://github.com/mhalder/qdrant-mcp-server/issues/32)
+* ci: add codecov configuration for coverage thresholds ([153a85e](https://github.com/mhalder/qdrant-mcp-server/commit/153a85e))
+* ci: make codecov project check informational ([f8a09c3](https://github.com/mhalder/qdrant-mcp-server/commit/f8a09c3))
+* test: add comprehensive test suite for code vectorization (802/840 passing) ([969c000](https://github.com/mhalder/qdrant-mcp-server/commit/969c000))
+* test: add path validation tests to improve coverage ([4cbf5ed](https://github.com/mhalder/qdrant-mcp-server/commit/4cbf5ed))
+* test: fix test samples and secret detection (809/840 passing) ([963a5c7](https://github.com/mhalder/qdrant-mcp-server/commit/963a5c7))
+* test: improve test coverage to 97.72% and fix race conditions ([78a75cd](https://github.com/mhalder/qdrant-mcp-server/commit/78a75cd))
+* feat: add code vectorization for semantic code search (#31) ([2a3745e](https://github.com/mhalder/qdrant-mcp-server/commit/2a3745e)), closes [#31](https://github.com/mhalder/qdrant-mcp-server/issues/31) [#31](https://github.com/mhalder/qdrant-mcp-server/issues/31)
+* feat: add hybrid search and advanced filtering for code search (#31) ([bfb022e](https://github.com/mhalder/qdrant-mcp-server/commit/bfb022e)), closes [#31](https://github.com/mhalder/qdrant-mcp-server/issues/31) [#31](https://github.com/mhalder/qdrant-mcp-server/issues/31)
+* feat: add incremental re-indexing with Merkle tree change detection (#31) ([dcd7c55](https://github.com/mhalder/qdrant-mcp-server/commit/dcd7c55)), closes [#31](https://github.com/mhalder/qdrant-mcp-server/issues/31) [Hi#level](https://github.com/Hi/issues/level) [#31](https://github.com/mhalder/qdrant-mcp-server/issues/31)
+* feat: add path traversal validation for security ([0d783cc](https://github.com/mhalder/qdrant-mcp-server/commit/0d783cc)), closes [#2](https://github.com/mhalder/qdrant-mcp-server/issues/2)
+* fix: improve test coverage and fix critical bugs (830/840 passing) ([32c13b2](https://github.com/mhalder/qdrant-mcp-server/commit/32c13b2))
+* fix: resolve tree-sitter peer dependency conflicts for Node.js 22.x ([61d614f](https://github.com/mhalder/qdrant-mcp-server/commit/61d614f))
+* docs: add comprehensive code vectorization examples and documentation ([1e0d48d](https://github.com/mhalder/qdrant-mcp-server/commit/1e0d48d))
+* style: format code vectorization module with project formatter ([beb71bb](https://github.com/mhalder/qdrant-mcp-server/commit/beb71bb))
+
+## 1.4.0 (2025-10-13)
+
+* Merge pull request #29 from mhalder/feature/configurable-prompts ([f9652b1](https://github.com/mhalder/qdrant-mcp-server/commit/f9652b1)), closes [#29](https://github.com/mhalder/qdrant-mcp-server/issues/29)
+* feat: add support for configurable MCP prompts ([e3cc6c2](https://github.com/mhalder/qdrant-mcp-server/commit/e3cc6c2)), closes [#28](https://github.com/mhalder/qdrant-mcp-server/issues/28)
+* feat: set default prompts.json path with auto-loading ([203694a](https://github.com/mhalder/qdrant-mcp-server/commit/203694a))
+* test: add comprehensive tests for prompts feature ([bb9ca66](https://github.com/mhalder/qdrant-mcp-server/commit/bb9ca66))
+
 ## <small>1.3.1 (2025-10-12)</small>
 
 * Merge pull request #27 from mhalder/fix/empty-responses ([30d33a7](https://github.com/mhalder/qdrant-mcp-server/commit/30d33a7)), closes [#27](https://github.com/mhalder/qdrant-mcp-server/issues/27)

package/README.md

@@ -9,9 +9,12 @@ A Model Context Protocol (MCP) server providing semantic search capabilities usi
 
 - **Zero Setup**: Works out of the box with Ollama - no API keys required
 - **Privacy-First**: Local embeddings and vector storage - data never leaves your machine
+- **Code Vectorization**: Intelligent codebase indexing with AST-aware chunking and semantic code search
 - **Multiple Providers**: Ollama (default), OpenAI, Cohere, and Voyage AI
 - **Hybrid Search**: Combine semantic and keyword search for better results
 - **Semantic Search**: Natural language search with metadata filtering
+- **Incremental Indexing**: Efficient updates - only re-index changed files
+- **Configurable Prompts**: Create custom prompts for guided workflows without code changes
 - **Rate Limiting**: Intelligent throttling with exponential backoff
 - **Full CRUD**: Create, search, and manage collections and documents
 - **Flexible Deployment**: Run locally (stdio) or as a remote HTTP server
@@ -123,11 +126,273 @@ See [Advanced Configuration](#advanced-configuration) section below for all opti
 | `hybrid_search`    | Hybrid search combining semantic and keyword (BM25) search with RRF           |
 | `delete_documents` | Delete specific documents by ID                                               |
 
+### Code Vectorization
+
+| Tool               | Description                                                                                    |
+| ------------------ | ---------------------------------------------------------------------------------------------- |
+| `index_codebase`   | Index a codebase for semantic code search with AST-aware chunking                              |
+| `search_code`      | Search indexed codebase using natural language queries                                         |
+| `reindex_changes`  | Incrementally re-index only changed files (detects added/modified/deleted)                     |
+| `get_index_status` | Get indexing status and statistics for a codebase                                              |
+| `clear_index`      | Delete all indexed data for a codebase                                                         |
+
 ### Resources
 
 - `qdrant://collections` - List all collections
 - `qdrant://collection/{name}` - Collection details
 
+## Configurable Prompts
+
+Create custom prompts tailored to your specific use cases without modifying code. Prompts provide guided workflows for common tasks.
+
+**Note**: By default, the server looks for `prompts.json` in the project root directory. If the file exists, prompts are automatically loaded. You can specify a custom path using the `PROMPTS_CONFIG_FILE` environment variable.
+
+### Setup
+
+1. **Create a prompts configuration file** (e.g., `prompts.json` in the project root):
+
+   See [`prompts.example.json`](prompts.example.json) for example configurations you can copy and customize.
+
+2. **Configure the server** (optional - only needed for custom path):
+
+If you place `prompts.json` in the project root, no additional configuration is needed. To use a custom path:
+
+```json
+{
+  "mcpServers": {
+    "qdrant": {
+      "command": "node",
+      "args": ["/path/to/qdrant-mcp-server/build/index.js"],
+      "env": {
+        "QDRANT_URL": "http://localhost:6333",
+        "PROMPTS_CONFIG_FILE": "/custom/path/to/prompts.json"
+      }
+    }
+  }
+}
+```
+
+3. **Use prompts** in your AI assistant:
+
+**Claude Code:**
+
+```bash
+/mcp__qdrant__find_similar_docs papers "neural networks" 10
+```
+
+**VSCode:**
+
+```bash
+/mcp.qdrant.find_similar_docs papers "neural networks" 10
+```
+
+### Example Prompts
+
+See [`prompts.example.json`](prompts.example.json) for ready-to-use prompts including:
+
+- `find_similar_docs` - Semantic search with result explanation
+- `setup_rag_collection` - Create RAG-optimized collections
+- `analyze_collection` - Collection insights and recommendations
+- `bulk_add_documents` - Guided bulk document insertion
+- `search_with_filter` - Metadata filtering assistance
+- `compare_search_methods` - Semantic vs hybrid search comparison
+- `collection_maintenance` - Maintenance and cleanup workflows
+- `migrate_to_hybrid` - Collection migration guide
+
+### Template Syntax
+
+Templates use `{{variable}}` placeholders:
+
+- Required arguments must be provided
+- Optional arguments use defaults if not specified
+- Unknown variables are left as-is in the output
+
+## Code Vectorization
+
+Intelligently index and search your codebase using semantic code search. Perfect for AI-assisted development, code exploration, and understanding large codebases.
+
+### Features
+
+- **AST-Aware Chunking**: Intelligent code splitting at function/class boundaries using tree-sitter
+- **Multi-Language Support**: 35+ file types including TypeScript, Python, Java, Go, Rust, C++, and more
+- **Incremental Updates**: Only re-index changed files for fast updates
+- **Smart Ignore Patterns**: Respects .gitignore, .dockerignore, and custom .contextignore files
+- **Semantic Search**: Natural language queries to find relevant code
+- **Metadata Filtering**: Filter by file type, path patterns, or language
+- **Local-First**: All processing happens locally - your code never leaves your machine
+
+### Quick Start
+
+**1. Index your codebase:**
+
+```bash
+# Via Claude Code MCP tool
+/mcp__qdrant__index_codebase /path/to/your/project
+```
+
+**2. Search your code:**
+
+```bash
+# Natural language search
+/mcp__qdrant__search_code /path/to/your/project "authentication middleware"
+
+# Filter by file type
+/mcp__qdrant__search_code /path/to/your/project "database schema" --fileTypes .ts,.js
+
+# Filter by path pattern
+/mcp__qdrant__search_code /path/to/your/project "API endpoints" --pathPattern src/api/**
+```
+
+**3. Update after changes:**
+
+```bash
+# Incrementally re-index only changed files
+/mcp__qdrant__reindex_changes /path/to/your/project
+```
+
+### Usage Examples
+
+#### Index a TypeScript Project
+
+```typescript
+// The MCP tool automatically:
+// 1. Scans all .ts, .tsx, .js, .jsx files
+// 2. Respects .gitignore patterns (skips node_modules, dist, etc.)
+// 3. Chunks code at function/class boundaries
+// 4. Generates embeddings using your configured provider
+// 5. Stores in Qdrant with metadata (file path, line numbers, language)
+
+index_codebase({
+  path: "/workspace/my-app",
+  forceReindex: false  // Set to true to re-index from scratch
+})
+
+// Output:
+// ✓ Indexed 247 files (1,823 chunks) in 45.2s
+```
+
+#### Search for Authentication Code
+
+```typescript
+search_code({
+  path: "/workspace/my-app",
+  query: "how does user authentication work?",
+  limit: 5
+})
+
+// Results include file path, line numbers, and code snippets:
+// [
+//   {
+//     filePath: "src/auth/middleware.ts",
+//     startLine: 15,
+//     endLine: 42,
+//     content: "export async function authenticateUser(req: Request) { ... }",
+//     score: 0.89,
+//     language: "typescript"
+//   },
+//   ...
+// ]
+```
+
+#### Search with Filters
+
+```typescript
+// Only search TypeScript files
+search_code({
+  path: "/workspace/my-app",
+  query: "error handling patterns",
+  fileTypes: [".ts", ".tsx"],
+  limit: 10
+})
+
+// Only search in specific directories
+search_code({
+  path: "/workspace/my-app",
+  query: "API route handlers",
+  pathPattern: "src/api/**",
+  limit: 10
+})
+```
+
+#### Incremental Re-indexing
+
+```typescript
+// After making changes to your codebase
+reindex_changes({
+  path: "/workspace/my-app"
+})
+
+// Output:
+// ✓ Updated: +3 files added, ~5 files modified, -1 files deleted
+// ✓ Chunks: +47 added, -23 deleted in 8.3s
+```
+
+#### Check Indexing Status
+
+```typescript
+get_index_status({
+  path: "/workspace/my-app"
+})
+
+// Output:
+// {
+//   isIndexed: true,
+//   collectionName: "code_a3f8d2e1",
+//   chunksCount: 1823,
+//   filesCount: 247,
+//   lastUpdated: "2025-01-30T10:15:00Z",
+//   languages: ["typescript", "javascript", "json"]
+// }
+```
+
+### Supported Languages
+
+**Programming Languages** (35+ file types):
+- **Web**: TypeScript, JavaScript, Vue, Svelte
+- **Backend**: Python, Java, Go, Rust, Ruby, PHP
+- **Systems**: C, C++, C#
+- **Mobile**: Swift, Kotlin, Dart
+- **Functional**: Scala, Clojure, Haskell, OCaml
+- **Scripting**: Bash, Shell, Fish
+- **Data**: SQL, GraphQL, Protocol Buffers
+- **Config**: JSON, YAML, TOML, XML, Markdown
+
+See [configuration](#code-vectorization-configuration) for full list and customization options.
+
+### Custom Ignore Patterns
+
+Create a `.contextignore` file in your project root to specify additional patterns to ignore:
+
+```gitignore
+# .contextignore
+**/test/**
+**/*.test.ts
+**/*.spec.ts
+**/fixtures/**
+**/mocks/**
+**/__tests__/**
+```
+
+### Best Practices
+
+1. **Index Once, Update Incrementally**: Use `index_codebase` for initial indexing, then `reindex_changes` for updates
+2. **Use Filters**: Narrow search scope with `fileTypes` and `pathPattern` for better results
+3. **Meaningful Queries**: Use natural language that describes what you're looking for (e.g., "database connection pooling" instead of "db")
+4. **Check Status First**: Use `get_index_status` to verify a codebase is indexed before searching
+5. **Local Embedding**: Use Ollama (default) to keep everything local and private
+
+### Performance
+
+Typical performance on a modern laptop (Apple M1/M2 or similar):
+
+| Codebase Size | Files | Indexing Time | Search Latency |
+|--------------|-------|---------------|----------------|
+| Small (10k LOC) | 50 | ~10s | <100ms |
+| Medium (100k LOC) | 500 | ~2min | <200ms |
+| Large (500k LOC) | 2,500 | ~10min | <500ms |
+
+**Note**: Indexing time varies based on embedding provider. Ollama (local) is fastest for initial indexing.
+
 ## Examples
 
 See [examples/](examples/) directory for detailed guides:
@@ -136,17 +401,26 @@ See [examples/](examples/) directory for detailed guides:
 - **[Knowledge Base](examples/knowledge-base/)** - Structured documentation with metadata
 - **[Advanced Filtering](examples/filters/)** - Complex boolean filters
 - **[Rate Limiting](examples/rate-limiting/)** - Batch processing with cloud providers
+- **[Code Search](examples/code-search/)** - Index codebases and semantic code search
 
 ## Advanced Configuration
 
 ### Environment Variables
 
+#### Core Configuration
+
 | Variable                            | Description                            | Default               |
 | ----------------------------------- | -------------------------------------- | --------------------- |
 | `TRANSPORT_MODE`                    | "stdio" or "http"                      | stdio                 |
 | `HTTP_PORT`                         | Port for HTTP transport                | 3000                  |
 | `EMBEDDING_PROVIDER`                | "ollama", "openai", "cohere", "voyage" | ollama                |
 | `QDRANT_URL`                        | Qdrant server URL                      | http://localhost:6333 |
+| `PROMPTS_CONFIG_FILE`               | Path to prompts configuration JSON     | prompts.json          |
+
+#### Embedding Configuration
+
+| Variable                            | Description                            | Default               |
+| ----------------------------------- | -------------------------------------- | --------------------- |
 | `EMBEDDING_MODEL`                   | Model name                             | Provider-specific     |
 | `EMBEDDING_BASE_URL`                | Custom API URL                         | Provider-specific     |
 | `EMBEDDING_MAX_REQUESTS_PER_MINUTE` | Rate limit                             | Provider-specific     |
@@ -156,6 +430,18 @@ See [examples/](examples/) directory for detailed guides:
 | `COHERE_API_KEY`                    | Cohere API key                         | -                     |
 | `VOYAGE_API_KEY`                    | Voyage AI API key                      | -                     |
 
+#### Code Vectorization Configuration
+
+| Variable                  | Description                                      | Default |
+| ------------------------- | ------------------------------------------------ | ------- |
+| `CODE_CHUNK_SIZE`         | Maximum chunk size in characters                 | 2500    |
+| `CODE_CHUNK_OVERLAP`      | Overlap between chunks in characters             | 300     |
+| `CODE_ENABLE_AST`         | Enable AST-aware chunking (tree-sitter)          | true    |
+| `CODE_BATCH_SIZE`         | Number of chunks to embed in one batch           | 100     |
+| `CODE_CUSTOM_EXTENSIONS`  | Additional file extensions (comma-separated)     | -       |
+| `CODE_CUSTOM_IGNORE`      | Additional ignore patterns (comma-separated)     | -       |
+| `CODE_DEFAULT_LIMIT`      | Default search result limit                      | 5       |
+
 ### Provider Comparison
 
 | Provider   | Models                                                          | Dimensions     | Rate Limit | Notes                |
@@ -169,15 +455,20 @@ See [examples/](examples/) directory for detailed guides:
 
 ## Troubleshooting
 
-| Issue                  | Solution                                                                     |
-| ---------------------- | ---------------------------------------------------------------------------- |
-| **Qdrant not running** | `docker compose up -d`                                                       |
-| **Collection missing** | Create collection first before adding documents                              |
-| **Ollama not running** | Verify with `curl http://localhost:11434`, start with `docker compose up -d` |
-| **Model missing**      | `docker exec ollama ollama pull nomic-embed-text`                            |
-| **Rate limit errors**  | Adjust `EMBEDDING_MAX_REQUESTS_PER_MINUTE` to match your provider tier       |
-| **API key errors**     | Verify correct API key in environment configuration                          |
-| **Filter errors**      | Ensure Qdrant filter format, check field names match metadata                |
+| Issue                           | Solution                                                                     |
+| ------------------------------- | ---------------------------------------------------------------------------- |
+| **Qdrant not running**          | `docker compose up -d`                                                       |
+| **Collection missing**          | Create collection first before adding documents                              |
+| **Ollama not running**          | Verify with `curl http://localhost:11434`, start with `docker compose up -d` |
+| **Model missing**               | `docker exec ollama ollama pull nomic-embed-text`                            |
+| **Rate limit errors**           | Adjust `EMBEDDING_MAX_REQUESTS_PER_MINUTE` to match your provider tier       |
+| **API key errors**              | Verify correct API key in environment configuration                          |
+| **Filter errors**               | Ensure Qdrant filter format, check field names match metadata                |
+| **Codebase not indexed**        | Run `index_codebase` before `search_code`                                    |
+| **Slow indexing**               | Use Ollama (local) for faster indexing, or increase `CODE_BATCH_SIZE`        |
+| **Files not found**             | Check `.gitignore` and `.contextignore` patterns                             |
+| **Search returns no results**   | Try broader queries, check if codebase is indexed with `get_index_status`    |
+| **Out of memory during index**  | Reduce `CODE_CHUNK_SIZE` or `CODE_BATCH_SIZE`                                |
 
 ## Development
 
@@ -208,6 +499,10 @@ Contributions welcome! See [CONTRIBUTING.md](CONTRIBUTING.md) for:
 
 **Automated releases**: Semantic versioning via conventional commits - `feat:` → minor, `fix:` → patch, `BREAKING CHANGE:` → major.
 
+## Acknowledgments
+
+The code vectorization feature is inspired by and builds upon concepts from the excellent [claude-context](https://github.com/zilliztech/claude-context) project (MIT License, Copyright 2025 Zilliz).
+
 ## License
 
 MIT - see [LICENSE](LICENSE) file.

package/build/code/chunker/base.d.ts

@@ -0,0 +1,19 @@
+/**
+ * Base interface for code chunkers
+ */
+import type { CodeChunk } from "../types.js";
+export interface CodeChunker {
+    /**
+     * Split code into semantic chunks
+     */
+    chunk(code: string, filePath: string, language: string): Promise<CodeChunk[]>;
+    /**
+     * Check if language is supported by this chunker
+     */
+    supportsLanguage(language: string): boolean;
+    /**
+     * Get chunking strategy name
+     */
+    getStrategyName(): string;
+}
+//# sourceMappingURL=base.d.ts.map

package/build/code/chunker/base.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"base.d.ts","sourceRoot":"","sources":["../../../src/code/chunker/base.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAE7C,MAAM,WAAW,WAAW;IAC1B;;OAEG;IACH,KAAK,CAAC,IAAI,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,SAAS,EAAE,CAAC,CAAC;IAE9E;;OAEG;IACH,gBAAgB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;IAE5C;;OAEG;IACH,eAAe,IAAI,MAAM,CAAC;CAC3B"}

package/build/code/chunker/base.js

@@ -0,0 +1,5 @@
+/**
+ * Base interface for code chunkers
+ */
+export {};
+//# sourceMappingURL=base.js.map

package/build/code/chunker/base.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"base.js","sourceRoot":"","sources":["../../../src/code/chunker/base.ts"],"names":[],"mappings":"AAAA;;GAEG"}

package/build/code/chunker/character-chunker.d.ts

@@ -0,0 +1,22 @@
+/**
+ * CharacterChunker - Simple character-based chunking with overlap
+ * Used as fallback when AST parsing is not available
+ */
+import type { ChunkerConfig, CodeChunk } from "../types.js";
+import type { CodeChunker } from "./base.js";
+export declare class CharacterChunker implements CodeChunker {
+    private config;
+    constructor(config: ChunkerConfig);
+    chunk(code: string, filePath: string, language: string): Promise<CodeChunk[]>;
+    supportsLanguage(_language: string): boolean;
+    getStrategyName(): string;
+    /**
+     * Find a good break point in the code (empty line, closing brace, etc.)
+     */
+    private findBreakPoint;
+    /**
+     * Calculate number of lines to overlap based on chunk size
+     */
+    private calculateOverlapLines;
+}
+//# sourceMappingURL=character-chunker.d.ts.map

package/build/code/chunker/character-chunker.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"character-chunker.d.ts","sourceRoot":"","sources":["../../../src/code/chunker/character-chunker.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,KAAK,EAAE,aAAa,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAC5D,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,WAAW,CAAC;AAE7C,qBAAa,gBAAiB,YAAW,WAAW;IACtC,OAAO,CAAC,MAAM;gBAAN,MAAM,EAAE,aAAa;IAEnC,KAAK,CAAC,IAAI,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,SAAS,EAAE,CAAC;IA6EnF,gBAAgB,CAAC,SAAS,EAAE,MAAM,GAAG,OAAO;IAK5C,eAAe,IAAI,MAAM;IAIzB;;OAEG;IACH,OAAO,CAAC,cAAc;IAsBtB;;OAEG;IACH,OAAO,CAAC,qBAAqB;CAK9B"}

package/build/code/chunker/character-chunker.js

@@ -0,0 +1,111 @@
+/**
+ * CharacterChunker - Simple character-based chunking with overlap
+ * Used as fallback when AST parsing is not available
+ */
+export class CharacterChunker {
+    config;
+    constructor(config) {
+        this.config = config;
+    }
+    async chunk(code, filePath, language) {
+        const chunks = [];
+        const lines = code.split("\n");
+        let currentChunk = "";
+        let currentStartLine = 1;
+        let currentLineCount = 0;
+        let chunkIndex = 0;
+        for (let i = 0; i < lines.length; i++) {
+            const line = lines[i];
+            currentChunk += `${line}\n`;
+            currentLineCount++;
+            // Check if we've reached chunk size
+            if (currentChunk.length >= this.config.chunkSize) {
+                // Try to find a good break point (end of function, class, or empty line)
+                const breakPoint = this.findBreakPoint(lines, i + 1);
+                if (breakPoint > i && breakPoint - i < 20) {
+                    // Include lines up to break point, but respect maxChunkSize
+                    for (let j = i + 1; j <= breakPoint && j < lines.length; j++) {
+                        const nextLine = `${lines[j]}\n`;
+                        // Stop if adding this line would exceed maxChunkSize
+                        if (currentChunk.length + nextLine.length > this.config.maxChunkSize) {
+                            break;
+                        }
+                        currentChunk += nextLine;
+                        currentLineCount++;
+                        i = j;
+                    }
+                }
+                // Create chunk
+                chunks.push({
+                    content: currentChunk.trim(),
+                    startLine: currentStartLine,
+                    endLine: currentStartLine + currentLineCount - 1,
+                    metadata: {
+                        filePath,
+                        language,
+                        chunkIndex,
+                        chunkType: "block",
+                    },
+                });
+                chunkIndex++;
+                // Calculate overlap
+                const overlapLines = this.calculateOverlapLines(currentLineCount);
+                const _overlapStart = Math.max(0, currentLineCount - overlapLines);
+                // Start new chunk with overlap
+                currentChunk = `${lines.slice(i - overlapLines + 1, i + 1).join("\n")}\n`;
+                currentStartLine = currentStartLine + currentLineCount - overlapLines;
+                currentLineCount = overlapLines;
+            }
+        }
+        // Add remaining content as final chunk
+        if (currentChunk.trim().length > 50) {
+            chunks.push({
+                content: currentChunk.trim(),
+                startLine: currentStartLine,
+                endLine: currentStartLine + currentLineCount - 1,
+                metadata: {
+                    filePath,
+                    language,
+                    chunkIndex,
+                    chunkType: "block",
+                },
+            });
+        }
+        return chunks;
+    }
+    supportsLanguage(_language) {
+        // Character chunker supports all languages
+        return true;
+    }
+    getStrategyName() {
+        return "character-based";
+    }
+    /**
+     * Find a good break point in the code (empty line, closing brace, etc.)
+     */
+    findBreakPoint(lines, startIdx) {
+        const searchWindow = Math.min(20, lines.length - startIdx);
+        for (let i = 0; i < searchWindow; i++) {
+            const line = lines[startIdx + i]?.trim() || "";
+            // Good break points
+            if (line === "" ||
+                line === "}" ||
+                line === "};" ||
+                line === "]);" ||
+                line.startsWith("//") ||
+                line.startsWith("#")) {
+                return startIdx + i;
+            }
+        }
+        return startIdx;
+    }
+    /**
+     * Calculate number of lines to overlap based on chunk size
+     */
+    calculateOverlapLines(totalLines) {
+        const overlapChars = this.config.chunkOverlap;
+        const avgCharsPerLine = this.config.chunkSize / Math.max(totalLines, 1);
+        return Math.floor(overlapChars / Math.max(avgCharsPerLine, 1));
+    }
+}
+//# sourceMappingURL=character-chunker.js.map

package/build/code/chunker/character-chunker.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"character-chunker.js","sourceRoot":"","sources":["../../../src/code/chunker/character-chunker.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAKH,MAAM,OAAO,gBAAgB;IACP;IAApB,YAAoB,MAAqB;QAArB,WAAM,GAAN,MAAM,CAAe;IAAG,CAAC;IAE7C,KAAK,CAAC,KAAK,CAAC,IAAY,EAAE,QAAgB,EAAE,QAAgB;QAC1D,MAAM,MAAM,GAAgB,EAAE,CAAC;QAC/B,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAE/B,IAAI,YAAY,GAAG,EAAE,CAAC;QACtB,IAAI,gBAAgB,GAAG,CAAC,CAAC;QACzB,IAAI,gBAAgB,GAAG,CAAC,CAAC;QACzB,IAAI,UAAU,GAAG,CAAC,CAAC;QAEnB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;YACtC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;YACtB,YAAY,IAAI,GAAG,IAAI,IAAI,CAAC;YAC5B,gBAAgB,EAAE,CAAC;YAEnB,oCAAoC;YACpC,IAAI,YAAY,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAC,SAAS,EAAE,CAAC;gBACjD,yEAAyE;gBACzE,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;gBAErD,IAAI,UAAU,GAAG,CAAC,IAAI,UAAU,GAAG,CAAC,GAAG,EAAE,EAAE,CAAC;oBAC1C,4DAA4D;oBAC5D,KAAK,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,UAAU,IAAI,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;wBAC7D,MAAM,QAAQ,GAAG,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC;wBACjC,qDAAqD;wBACrD,IAAI,YAAY,CAAC,MAAM,GAAG,QAAQ,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,YAAY,EAAE,CAAC;4BACrE,MAAM;wBACR,CAAC;wBACD,YAAY,IAAI,QAAQ,CAAC;wBACzB,gBAAgB,EAAE,CAAC;wBACnB,CAAC,GAAG,CAAC,CAAC;oBACR,CAAC;gBACH,CAAC;gBAED,eAAe;gBACf,MAAM,CAAC,IAAI,CAAC;oBACV,OAAO,EAAE,YAAY,CAAC,IAAI,EAAE;oBAC5B,SAAS,EAAE,gBAAgB;oBAC3B,OAAO,EAAE,gBAAgB,GAAG,gBAAgB,GAAG,CAAC;oBAChD,QAAQ,EAAE;wBACR,QAAQ;wBACR,QAAQ;wBACR,UAAU;wBACV,SAAS,EAAE,OAAO;qBACnB;iBACF,CAAC,CAAC;gBAEH,UAAU,EAAE,CAAC;gBAEb,oBAAoB;gBACpB,MAAM,YAAY,GAAG,IAAI,CAAC,qBAAqB,CAAC,gBAAgB,CAAC,CAAC;gBAClE,MAAM,aAAa,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,gBAAgB,GAAG,YAAY,CAAC,CAAC;gBAEnE,+BAA+B;gBAC/B,YAAY,GAAG,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,GAAG,YAAY,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC1E,gBAAgB,GAAG,gBAAgB,GAAG,gBAAgB,GAAG,YAAY,CAAC;gBACtE,gBAAgB,GAAG,YAAY,CAAC;YAClC,CAAC;QACH,CAAC;QAED,uCAAuC;QACvC,IAAI,YAAY,CAAC,IAAI,EAAE,CAAC,MAAM,GAAG,EAAE,EAAE,CAAC;YACpC,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,YAAY,CAAC,IAAI,EAAE;gBAC5B,SAAS,EAAE,gBAAgB;gBAC3B,OAAO,EAAE,gBAAgB,GAAG,gBAAgB,GAAG,CAAC;gBAChD,QAAQ,EAAE;oBACR,QAAQ;oBACR,QAAQ;oBACR,UAAU;oBACV,SAAS,EAAE,OAAO;iBACnB;aACF,CAAC,CAAC;QACL,CAAC;QAED,OAAO,MAAM,CAAC;IAChB,CAAC;IAED,gBAAgB,CAAC,SAAiB;QAChC,2CAA2C;QAC3C,OAAO,IAAI,CAAC;IACd,CAAC;IAED,eAAe;QACb,OAAO,iBAAiB,CAAC;IAC3B,CAAC;IAED;;OAEG;IACK,cAAc,CAAC,KAAe,EAAE,QAAgB;QACtD,MAAM,YAAY,GAAG,IAAI,CAAC,GAAG,CAAC,EAAE,EAAE,KAAK,CAAC,MAAM,GAAG,QAAQ,CAAC,CAAC;QAE3D,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,YAAY,EAAE,CAAC,EAAE,EAAE,CAAC;YACtC,MAAM,IAAI,GAAG,KAAK,CAAC,QAAQ,GAAG,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,CAAC;YAE/C,oBAAoB;YACpB,IACE,IAAI,KAAK,EAAE;gBACX,IAAI,KAAK,GAAG;gBACZ,IAAI,KAAK,IAAI;gBACb,IAAI,KAAK,KAAK;gBACd,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC;gBACrB,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,EACpB,CAAC;gBACD,OAAO,QAAQ,GAAG,CAAC,CAAC;YACtB,CAAC;QACH,CAAC;QAED,OAAO,QAAQ,CAAC;IAClB,CAAC;IAED;;OAEG;IACK,qBAAqB,CAAC,UAAkB;QAC9C,MAAM,YAAY,GAAG,IAAI,CAAC,MAAM,CAAC,YAAY,CAAC;QAC9C,MAAM,eAAe,GAAG,IAAI,CAAC,MAAM,CAAC,SAAS,GAAG,IAAI,CAAC,GAAG,CAAC,UAAU,EAAE,CAAC,CAAC,CAAC;QACxE,OAAO,IAAI,CAAC,KAAK,CAAC,YAAY,GAAG,IAAI,CAAC,GAAG,CAAC,eAAe,EAAE,CAAC,CAAC,CAAC,CAAC;IACjE,CAAC;CACF"}

package/build/code/chunker/tree-sitter-chunker.d.ts

@@ -0,0 +1,29 @@
+/**
+ * TreeSitterChunker - AST-aware code chunking using tree-sitter
+ * Primary chunking strategy for supported languages
+ */
+import type { ChunkerConfig, CodeChunk } from "../types.js";
+import type { CodeChunker } from "./base.js";
+export declare class TreeSitterChunker implements CodeChunker {
+    private config;
+    private languages;
+    private fallbackChunker;
+    constructor(config: ChunkerConfig);
+    private initializeParsers;
+    chunk(code: string, filePath: string, language: string): Promise<CodeChunk[]>;
+    supportsLanguage(language: string): boolean;
+    getStrategyName(): string;
+    /**
+     * Find all chunkable nodes in the AST
+     */
+    private findChunkableNodes;
+    /**
+     * Extract function/class name from AST node
+     */
+    private extractName;
+    /**
+     * Map AST node type to chunk type
+     */
+    private getChunkType;
+}
+//# sourceMappingURL=tree-sitter-chunker.d.ts.map

package/build/code/chunker/tree-sitter-chunker.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"tree-sitter-chunker.d.ts","sourceRoot":"","sources":["../../../src/code/chunker/tree-sitter-chunker.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAYH,OAAO,KAAK,EAAE,aAAa,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAC5D,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,WAAW,CAAC;AAQ7C,qBAAa,iBAAkB,YAAW,WAAW;IAIvC,OAAO,CAAC,MAAM;IAH1B,OAAO,CAAC,SAAS,CAA0C;IAC3D,OAAO,CAAC,eAAe,CAAmB;gBAEtB,MAAM,EAAE,aAAa;IAKzC,OAAO,CAAC,iBAAiB;IAgFnB,KAAK,CAAC,IAAI,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,SAAS,EAAE,CAAC;IAoEnF,gBAAgB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO;IAI3C,eAAe,IAAI,MAAM;IAIzB;;OAEG;IACH,OAAO,CAAC,kBAAkB;IAsB1B;;OAEG;IACH,OAAO,CAAC,WAAW;IAiBnB;;OAEG;IACH,OAAO,CAAC,YAAY;CAYrB"}