tf2-wiki-mcp
Gives LLMs live, accurate access to Team Fortress Wiki content, enabling search, page retrieval, structured weapon stats, patch notes, and other TF2-specific information.
README
tf2-wiki-mcp
An MCP server that exposes the Team Fortress Wiki to LLM clients (Claude Desktop, Claude Code, Cursor, VS Code) over stdio.
Not affiliated with Valve Corporation or the Team Fortress Wiki contributors. Team Fortress 2 is a trademark of Valve Corporation. This is unofficial fan tooling.
What it does
Gives an LLM live, accurate access to TF2 wiki content so it stops guessing at weapon stats, patch notes, and cosmetic details.
- Generic wiki access — search, fetch pages, get summaries, list recent changes.
- TF2 domain tools — structured weapon stats, class loadouts, cosmetic lookup, event item lists, patch notes.
Tools
| Tool | Purpose |
|---|---|
search_wiki(query, limit) |
Full-text search |
get_page(title, format) |
Page content as wikitext, plain, or html |
get_page_summary(title) |
Lead-section plaintext |
get_page_sections(title) |
Section TOC for selective fetching |
get_recent_changes(limit, namespace) |
Recently edited pages |
get_weapon_stats(weapon_name) |
Parsed weapon infobox → structured stats |
list_class_loadout(class_name) |
All weapons in a class's wiki category |
get_cosmetic(name) |
Cosmetic item infobox params |
list_event_items(event_name) |
Items added in an update/event |
get_patch_notes(update_name) |
Patch notes for a specific update |
Plus: resource tf2wiki://main, prompts analyze_loadout and compare_weapons.
Installation
Requires uv (which manages the Python toolchain for you). Once installed, uvx tf2-wiki-mcp will fetch and run the server on demand — no manual clone needed.
Standard config works in most MCP clients:
{
"mcpServers": {
"tf2-wiki-mcp": {
"command": "uvx",
"args": ["tf2-wiki-mcp"],
"env": {}
}
}
}
<details> <summary>Claude Desktop</summary>
Add the standard config above to claude_desktop_config.json:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Restart Claude Desktop. The TF2 wiki tools appear in the tool picker. </details>
<details> <summary>Claude Code</summary>
claude mcp add tf2-wiki-mcp -- uvx tf2-wiki-mcp
</details>
<details> <summary>Codex</summary>
Add to ~/.codex/config.toml:
[mcp_servers.tf2-wiki-mcp]
command = "uvx"
args = ["tf2-wiki-mcp"]
See the Codex MCP docs. </details>
<details> <summary>Cursor</summary>
Or add the standard config to ~/.cursor/mcp.json.
</details>
<details> <summary>Gemini CLI</summary>
Add the standard config to your Gemini CLI settings.json per the Gemini CLI MCP guide.
</details>
<details> <summary>VS Code</summary>
Or via the VS Code CLI:
code --add-mcp '{"name":"tf2-wiki-mcp","command":"uvx","args":["tf2-wiki-mcp"],"env":{}}'
See the VS Code MCP guide for details. </details>
<details> <summary>VS Code Insiders</summary>
Or via the CLI:
code-insiders --add-mcp '{"name":"tf2-wiki-mcp","command":"uvx","args":["tf2-wiki-mcp"],"env":{}}'
</details>
<details> <summary>Visual Studio</summary>
Or manually:
- Open the GitHub Copilot Chat window.
- Click the tools icon (🛠️) → + Add Server.
- Fill in: Server ID
tf2-wiki-mcp, Typestdio, Commanduvx, Argumentstf2-wiki-mcp. - Save.
See the Visual Studio MCP docs. </details>
<details> <summary>GitHub Copilot Coding Agent</summary>
{
"mcpServers": {
"tf2-wiki-mcp": {
"command": "uvx",
"args": ["tf2-wiki-mcp"],
"env": {},
"type": "local",
"tools": ["*"]
}
}
}
Add this in repository settings under Copilot → Coding agent. See the Copilot Coding Agent MCP docs. </details>
Development
Run from a local clone instead of PyPI:
git clone https://github.com/yusufaf/tf2-wiki-mcp
cd tf2-wiki-mcp
uv sync --dev
uv run pytest # offline, uses recorded cassettes
uv run pytest --live # also hits the real wiki
To point an MCP client at the local checkout:
{
"mcpServers": {
"tf2-wiki-mcp": {
"command": "uv",
"args": ["run", "--directory", "/absolute/path/to/tf2-wiki-mcp", "tf2-wiki-mcp"]
}
}
}
Licensing & Attribution
Code: MIT (see LICENSE).
Wiki content: Not redistributed. This project contains zero scraped wiki data. All page content is fetched live from wiki.teamfortress.com at the user's request and returned directly to the user's LLM client — the same posture as a browser extension.
Wiki content is © its respective contributors under Valve's Steam Subscriber Agreement (Game Site terms). Users of this tool are responsible for compliance with Valve's terms. Code license ≠ content license: MIT covers this codebase, not the wiki content it fetches.
Requests to the wiki include a descriptive User-Agent (tf2-wiki-mcp/<version> (https://github.com/yusufaf/tf2-wiki-mcp)) per MediaWiki etiquette and respect maxlag/Retry-After responses.
Contributing
Don't check in scraped wiki text. Infobox examples for parser tests are fine (short, fair-use-grade fixtures). Full page dumps are not.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.