Happy slice of mcp pizzaMCP.pizza

FACT

MCP.Pizza Chef: ruvnet

Rather than guessing from a pile of similar-looking documents, this runs the precise lookup your question calls for, such as a database query or a live service call, and hands back the real numbers. Background information that never changes is cached, so repeat questions come back fast and cost less. Data access is read-only and every lookup is logged. This is a Python project aimed at developers, so expect to install it yourself and supply an Anthropic key.

Coding
Data

Use This MCP server To

Ask for last quarter's sales numbers and get exact figures Check live inventory levels without opening a dashboard Trace every answer back to the lookup that produced it Cut the cost of repeated questions over the same data Replace a slow document search with exact database answers

README

FACT: Fast Augmented Context Tools

A revolutionary approach to LLM data retrieval that replaces RAG with prompt caching and deterministic tool execution under the Model Context Protocol

TL;DR

FACT (Fast Augmented Context Tools) introduces a new paradigm for language model–powered data retrieval by replacing vector-based retrieval with a prompt-and-tool approach under the Model Context Protocol (MCP). The result? Sub-100ms responses, 60-90% cost reduction, and deterministic, auditable results with no vector stores required.

Why FACT? RAG Had Its Moment. It's Time for Something Smarter.

RAG (Retrieval-Augmented Generation) made sense when vector search was the best we had. But vectors are slow, fuzzy, and expensive to maintain. They're inherently imprecise, forcing you to tune similarity thresholds, re-embed documents, and accept that relevance is always a bit of a guess.

What we needed was something explicit. Deterministic. Cheap. Fast.

FACT isn't about fetching similar chunks of data. It's about giving models structured, exact answers via tool execution and pairing that with intelligent prompt caching. Prompt caches work like brains with memory. Tools act like hands that do. And when you combine the two—prompt caching + MCP-based tools—you can skip vector search entirely.

Instead of saying "Find me something like this," FACT says: "Run this exact SQL call. Return this live API result. Use this schema. Cache the output."

Introduction to FACT

FACT (Fast Augmented Context Tools) introduces a new paradigm for language model–powered data retrieval by replacing vector-based retrieval with a prompt-and-tool approach under the Model Context Protocol (MCP). Instead of relying on embeddings and similarity searches, FACT combines intelligent prompt caching with deterministic tool invocation to deliver fresh, precise, and auditable results.

Key Differences from RAG

FACT represents a fundamental shift from traditional RAG (Retrieval-Augmented Generation) approaches:

Retrieval Mechanism

FACT FAQ

Who is this for?
Developers building a question-answering setup over their own data. It is not a click-to-install add-on.
Do I need a key?
Yes, an Anthropic key is required, and an Arcade key is optional for the hosted tools.
Can I use this to ask questions about my own company data?
Yes, questions are answered by running real lookups against the data sources you connect to it.
How hard is setup?
Technical. Expect Python 3.8 or newer, a configuration step, and your own data connection.
Can it change or delete my data?
No, access is read-only by design, and every operation is written to a log.
Does it need the usual search index setup?
No, it skips similarity search entirely and runs exact lookups instead.