pdf-brain 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Joel Hooks
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,137 @@
+# pdf-library
+
+Local PDF knowledge base with vector search. Extract, embed, and semantically search your PDFs.
+
+## Features
+
+- **Local-first** - Everything runs on your machine, no API costs
+- **Vector search** - Semantic search via Ollama embeddings (mxbai-embed-large)
+- **Hybrid search** - Combine vector similarity with full-text search
+- **iCloud sync** - Default storage in `~/Documents/.pdf-library/`
+- **Fast** - SQLite + sqlite-vec for instant queries
+
+## Requirements
+
+- macOS (Apple Silicon recommended)
+- [Bun](https://bun.sh) runtime
+- [Ollama](https://ollama.ai) for embeddings
+- [uv](https://github.com/astral-sh/uv) for PDF extraction
+
+## Setup
+
+```bash
+# Clone and setup
+git clone https://github.com/joelhooks/pdf-library.git
+cd pdf-library
+./scripts/setup.sh
+```
+
+This will:
+
+1. Install Ollama if needed
+2. Pull the `mxbai-embed-large` embedding model
+3. Install dependencies
+4. Create the library directory
+
+## Usage
+
+### CLI
+
+```bash
+# Add a PDF from local path
+bun run dev add /path/to/document.pdf
+
+# Add from URL
+bun run dev add https://example.com/paper.pdf
+
+# Add with tags
+bun run dev add /path/to/document.pdf --tags "ai,agents"
+
+# Add from URL with custom title
+bun run dev add https://example.com/paper.pdf --title "Research Paper" --tags "research"
+
+# Search semantically
+bun run dev search "context engineering patterns"
+
+# Full-text search
+bun run dev search "context engineering" --fts
+
+# List all documents
+bun run dev list
+
+# List by tag
+bun run dev list --tag ai
+
+# Get document details
+bun run dev get "document-title"
+
+# Remove a document
+bun run dev remove "document-title"
+
+# Update tags
+bun run dev tag "document-title" "new,tags,here"
+
+# Show stats
+bun run dev stats
+
+# Check Ollama status
+bun run dev check
+```
+
+### As a library
+
+```typescript
+import { PDFLibrary } from "pdf-library";
+
+const library = new PDFLibrary();
+
+// Add a PDF
+await library.add("/path/to/document.pdf", {
+  tags: ["ai", "agents"],
+});
+
+// Semantic search
+const results = await library.search("context engineering patterns");
+
+// Hybrid search (vector + FTS)
+const results = await library.search("context engineering", { hybrid: true });
+
+// List documents
+const docs = library.list();
+```
+
+### OpenCode Tool
+
+Copy `opencode-tool.ts` to `~/.config/opencode/tool/pdf-library.ts` to use as an OpenCode custom tool.
+
+## Configuration
+
+Environment variables:
+
+| Variable           | Default                    | Description              |
+| ------------------ | -------------------------- | ------------------------ |
+| `PDF_LIBRARY_PATH` | `~/Documents/.pdf-library` | Library storage location |
+| `OLLAMA_HOST`      | `http://localhost:11434`   | Ollama API endpoint      |
+| `OLLAMA_MODEL`     | `mxbai-embed-large`        | Embedding model          |
+
+## How it works
+
+1. **Extract** - PDF text extracted via `pypdf` (run through `uv`)
+2. **Chunk** - Text split into ~512 token chunks with overlap
+3. **Embed** - Each chunk embedded via Ollama (mxbai-embed-large, 1024 dims)
+4. **Store** - SQLite database with sqlite-vec for vector search + FTS5 for full-text
+5. **Search** - Query embedded, compared against chunks via cosine similarity
+
+## Storage
+
+```
+~/Documents/.pdf-library/
+├── library.db          # SQLite database (vectors, FTS, metadata)
+├── downloads/          # PDFs downloaded from URLs
+├── extracted/          # Markdown versions of PDFs (optional)
+└── originals/          # Copy of original PDFs (optional)
+```
+
+## License
+
+MIT