Archive Agent

Provides an MCP server to manage Archive Agent indexing, search, and RAG-based answers with local Qdrant storage.

python

57

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "shredengineer-archive-agent": {
      "command": "archive-agent",
      "args": [
        "mcp"
      ]
    }
  }
}

Archive Agent exposes an MCP server that lets you control and query its document indexing, RAG search, and local Qdrant storage from your MCP client. You can drive indexing, search, and answers through the MCP interface, enabling integration with IDEs, automation scripts, and other tools that support MCP protocol.

How to use

You connect to the Archive Agent MCP server from your MCP client and perform actions to track files, update the local index, search across your documents, and request answers generated by the RAG engine. Use the MCP server to manage a single or multiple profiles, run queries, and retrieve structured answers that reference the source chunks.

How to install

Prerequisites you need before installation:

Docker (for running Qdrant server)
Python >= 3.10

Concrete installation steps you should follow in a terminal in the directory where you want Archive Agent installed:

git clone https://github.com/shredEngineer/Archive-Agent
cd Archive-Agent
chmod +x install.sh
./install.sh

Additional content

Post-installation, you will configure an AI provider and then start the MCP server to enable MCP-based interactions.

Available tools

get_patterns

Retrieve the list of included and excluded patterns that determine which files Archive Agent tracks.

get_files_tracked

Return the list of files currently tracked by Archive Agent.

get_files_changed

Return the list of files that have changed since the last track or commit.

get_search_result

Get a list of files relevant to a given question or query.

get_answer_rag

Obtain an answer to a question using Retrieval Augmented Generation (RAG) with the retrieved context.