Doc qa
MCP-native framework for composable agent skills — typed tools from type hints, per-skill evals, runs fully offline
npx -y skills add Ninadnj/mcp-skills-kit --skill doc_qaAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Answer questions grounded in a set of documents (RAG), with citations.
SKILL.md
0.8 KB, 147 tokens by cl100k_base, as published. Nobody here has run it
doc_qa
Retrieval-augmented question answering over a folder of documents.
How it works
- Load
.txt/.md/.pdffiles from the given path. - Chunk them with overlap and rank chunks against the question (BM25).
- Answer only from the retrieved passages and cite the source file for each claim. If the answer isn't in the documents, say so.
Guidance for the model
- Ground every statement in the retrieved context; never invent facts.
- Keep answers concise and include the
[source]for each fact. - If the context does not contain the answer, reply that it is not in the documents.