https://github.com/jkingsman/qanon-mcp-server
## QAnon is a dangerous cult. This archive is for research purposes only, and I do _not_ endorse any material in this repo.
# Q-Anon Posts/Drops MCP Server
[](https://smithery.ai/server/@jkingsman/qanon-mcp-server)
An MCP (Model Context Protocol) server that provides access to a dataset of Q-Anon posts for anthropological/sociological research. This server allows AI assistants like Claude to search, filter, and analyze the Q-Anon drops.
Posts are drawn from https://github.com/jkingsman/JSON-QAnon. You can learn more about how the source data was composed there, as well as find alternate formats, schemas, etc.
### Warning: This tool was entirely vibe coded. Use at your own risk.
## Prerequisites
- Python 3.10 or higher
- `uv` package manager
- Claude Desktop (for Claude integration)
## Installation
This tool is compatible with `uvx` and doesn't need to be cloned/installed.
### Installing via Smithery
To install qanon-mcp-server for Claude Desktop automatically via [Smithery](https://smithery.ai/server/@jkingsman/qanon-mcp-server):
```bash
npx -y @smithery/cli install @jkingsman/qanon-mcp-server --client claude
```
### Manual
1. Clone or download this repository to your local machine
2. Install the required packages using `uv`:
```bash
uv pip install -e .
```
## Usage
You can run the server directly with `uvx`:
```bash
uvx qanon_mcp
```
## Claude Desktop Integration
To use this MCP server with Claude Desktop:
1. Make sure you have [Claude Desktop](https://claude.ai/download) installed
2. Open the Claude menu and select "Settings..."
3. Click on "Developer" in the left-hand bar and then "Edit Config"
4. Add the following configuration to the `claude_desktop_config.json` file:
```json
{
"mcpServers": {
"qanon_mcp": {
"command": "uvx",
"args": [
"qanon_mcp"
]
}
}
}
```
or, if you don't have `uvx` installed:
```json
{
"mcpServers": {
"qanon_mcp": {
"command": "uv",
"args": [
"tool",
"run",
"qanon_mcp"
]
}
}
}
```
5. Save the file and restart Claude Desktop
6. Start a new conversation in Claude Desktop
7. You should see a hammer icon in the input box, indicating that tools are available
## Features
### Resources
- `qanon://posts/count` - Get the total number of posts
- `qanon://posts/{post_id}` - Access a specific post by ID
- `qanon://posts/raw/{post_id}` - Get the raw JSON data for a specific post
- `qanon://authors` - List all unique authors
- `qanon://stats` - Get dataset statistics
### Tools
- **get_post_by_id_tool** - Retrieve a specific post by its ID
- **search_posts** - Find posts containing specific keywords or phrases
- **get_posts_by_date** - Retrieve posts from a specific date range
- **get_posts_by_author_id** - Find posts by a specific author ID
- **analyze_post** - Get detailed analysis of a specific post including references and context
- **get_timeline_summary** - Generate a chronological timeline, optionally within a date range
- **word_cloud_by_post_ids** - Generate a word frequency analysis for posts within a specified ID range
- **word_cloud_by_date_range** - Generate a word frequency analysis for posts within a specified date range
## Example Queries for Claude
Once the MCP server is connected to Claude Desktop, you can ask questions like:
- "How many Q-Anon posts are in the dataset?"
- "Search for posts that mention 'storm'"
- "Show me posts from October 2020"
- "Analyze post #3725"
- "Create a timeline of Q-Anon posts from 2018"
- "Generate a word cloud for Q-Anon posts between January and March 2019"
- "Get the raw data for post #4500"
- "What are the most common words used in posts #1000-2000?"
## Troubleshooting
- If Claude Desktop doesn't show the hammer icon, check your configuration and restart Claude Desktop
- Ensure the `posts.json` file is in the same directory as the script
- Check the output in the terminal for any error messages
- Make sure you're using the absolute path to the script in your Claude Desktop configuration
TDQS
Scored across 8 tools
Most tools have distinct purposes: retrieval by ID, author, date, search, timeline summary, and word cloud generation. However, 'analyze_post' and 'get_post_by_id_tool' could cause confusion as both target specific posts, with 'analyze_post' implying deeper analysis but overlapping in core retrieval functionality. The word cloud tools are clearly differentiated by date vs. ID range.
The naming is mixed: some tools use verb_noun patterns like 'search_posts' and 'analyze_post', while others use noun_verb patterns like 'get_post_by_id_tool' and 'get_posts_by_author_id'. There's inconsistency in verb usage ('get', 'analyze', 'search') and structure, but names remain readable and descriptive overall.
With 8 tools, this server is well-scoped for analyzing and retrieving posts/drops in a QAnon context. The count is reasonable, covering key operations like retrieval, search, timeline summary, and word cloud analysis without being overwhelming or too sparse.
The tool surface is largely complete for a read-only analysis server, covering retrieval by various criteria, search, timeline summaries, and word cloud generation. A minor gap is the lack of update or delete operations, but given the domain (likely analyzing historical data), this is acceptable, and agents can work around it effectively.