butterfly/docrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
23 Commits 2 Branches 0 Tags 1.4 MiB
main
Commit Graph

5 Commits

Author SHA1 Message Date
Z User
c03bde8023 Fix tool call parsing, improve embeddings, and fix async issues 
- main.py: Rewrote _parse_tool_call with brace-counting for robust JSON extraction
- main.py: Improved _clean_tool_syntax with brace-aware removal of tool_call JSON
- main.py: Fixed dict key mismatches (chunks_ingested, pages_downloaded)
- main.py: Run tool execution in asyncio.to_thread to avoid blocking event loop
- main.py: Always clean tool syntax from responses (handles edge cases)
- rag/__init__.py: Wrap blocking website_downloader in run_in_executor
- rag/__init__.py: Replace deprecated datetime.utcnow() with datetime.now(timezone.utc)
- rag/__init__.py: Add add_document_from_url method
- rag/vector_store.py: Replace hash-based embeddings with TF-IDF inspired embeddings
- rag/vector_store.py: Add embedding dimension mismatch handling in search
- README.md: Update API key config documentation
 2026-03-29 17:49:32 +00:00
Z User
b811162f78 Implement tool calling loop for LLM 
- Pass all registered tools to LLM during chat completion
- Handle tool_calls from LLM response
- Execute tools and feed results back to LLM
- Loop until LLM returns final response
- Updated system prompt to encourage tool use
- Updated streaming to handle tool calls
- Increased MAX_TOOL_ITERATIONS to 5
 2026-03-29 16:07:56 +00:00
Z User
eabdadfb62 Implement full DocRAG server with OpenAI-compatible API 
Features:
- FastAPI server with OpenAI-compatible endpoints (/v1/chat/completions, /v1/models)
- RAG system with document processing and vector storage
- Support for multiple document formats (PDF, DOCX, HTML, text, code)
- Streaming response support
- Tool integration with website_downloader
- Document management API endpoints
- GLM-4.7-Flash integration via z-ai-web-dev-sdk
- Works transparently with Open WebUI and other OpenAI clients

Components:
- main.py: FastAPI application with OpenAI-compatible API
- rag/: RAG system (document processor, vector store, retriever)
- tools/: Tool manager with website_downloader integration
- .env.example: Configuration template
 2026-03-29 00:57:37 +00:00
Z User
aa69b2f496 Add website downloader tool wrapper for GLM-4.7-Flash 
- Create website_downloader_tool.py with OpenAI function calling schema
- Add comprehensive tool documentation
- Update README with usage examples
- Update requirements.txt with optional sdk dependency
 2026-03-29 00:16:54 +00:00
butterfly
c02c032c67 Initial commit 2026-03-28 15:51:14 -07:00