Skip to content

Conversation

@haz3141
Copy link
Owner

@haz3141 haz3141 commented Sep 6, 2025

Implements Step 5 minimal RAG pipeline with:

  • Ingestion: - loads text files from data directory
  • Chunking: - simple sentence-based chunking with configurable max chars
  • Retrieval: - BM25 retriever using rank-bm25 library
  • MCP Endpoint: - exposes RAG functionality via HTTP API
  • Tests: Deterministic unit tests with no external dependencies
  • CI: Updated workflow to run RAG tests alongside existing tests
  • Config: Environment variables for data directory and top-k results

Dependencies: Only adds - lightweight and deterministic
Testing: Uses repo fixtures, no network calls, fully deterministic
Configuration: and environment variables

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants