iitc-mcp
Bridge between IITC and MCP agents, enabling AI assistants to interact with Ingress Intel: view map, count portals, track players, and communicate via COMM without manual input.
README
IITC-MCP
Bridge between IITC (Ingress Intel Total Conversion) and MCP (Model Context Protocol) agents. Let your AI assistant see the map, count portals, track players, and talk to COMM — without touching a mouse.

Quick Start
1. Install Userscript
Install IITC first, then add the iitc-mcp userscript in Tampermonkey:
https://github.com/comicchang/iitc-mcp/releases/latest/download/iitc-mcp.user.js
2. Configure MCP Server for Your Agent
CLI entry:
npx github:comicchang/iitc-mcp serve
Add --read-only for read-only mode (14 tools; omits iitc_send_comm and iitc_redeem_code):
npx github:comicchang/iitc-mcp serve --read-only
Codex (~/.codex/config.toml or project-level .codex/config.toml):
[mcp_servers.iitc-mcp]
command = "npx"
args = ["github:comicchang/iitc-mcp", "serve"]
OpenCode (~/.openCode/mcp.json or project-level .openCode/mcp.json):
{
"mcpServers": {
"iitc-mcp": {
"command": "npx",
"args": ["github:comicchang/iitc-mcp", "serve"]
}
}
}
<details> <summary>Oh My Pi local dev config</summary>
"iitc-mcp": {
"type": "stdio",
"command": "/path/to/node_modules/.bin/tsx",
"args": ["/path/to/packages/mcp-server/src/cli.ts", "serve"]
}
</details>
Reload MCP config and you're set — 16 tools auto-register (14 if server started with --read-only).
Open https://intel.ingress.com. Once both the userscript and MCP server are ready, the MCP indicator in IITC Toolbox turns green.
MCP Tools (16 total)
| Tool | Description |
|---|---|
iitc_get_map_state |
Map center, zoom, bounds, selected portal |
iitc_set_map_view |
Set map center and zoom |
iitc_fit_map_bounds |
Fit map to bounding box |
iitc_search_region |
Search named region via Nominatim → fit bounds → wait for data |
iitc_list_portals |
List portals in viewport (paginated) |
iitc_list_links |
List links in viewport (paginated) |
iitc_list_fields |
List control fields in viewport (paginated) |
iitc_get_portal_details |
Portal detail: mods, resonators, link/field GUIDs |
iitc_select_portal |
Select a portal on the map |
iitc_search |
Search portals by name |
iitc_list_comm |
Read COMM messages |
iitc_send_comm |
Send COMM message |
iitc_redeem_code |
Redeem a passcode |
iitc_get_self |
Your own faction, level, AP, XM |
iitc_list_players |
Tracked players with last position (Player Tracker) |
iitc_get_player_trail |
Single player's trail with timestamps |
Usage Examples
Ask your AI assistant in natural language:
Search a region and count portals
搜索静安雕塑公园,统计 portal 状态
iitc_search_region("静安雕塑公园") → 围框 + 等数据加载
iitc_list_portals → 按阵营统绿/蓝/红/白数量
Check a specific portal
青果巷赵宅现在什么颜色,连满 link 了吗
iitc_search("青果巷") → 找到候选 Portal
iitc_get_portal_details(guid) → 阵营/等级/血量/linkGuids
See who's been active nearby
附近最近有谁在动
iitc_list_players → 玩家名/阵营/最近位置/动作
iitc_get_player_trail("playerName") → 完整轨迹
Find high-value targets
区域内有哪些 L7+ Portal,哪些阵营占领的
iitc_search_region("目标区域") → 围框
iitc_list_portals → 按 level 筛选 L7+
Monitor COMM
看看 COMM 最近在聊什么
iitc_list_comm(channel="all") → 最近消息
Architecture
Default mode — embedded broker (one agent = one browser session):
graph LR
US[Userscript] -->|HTTP :27342| S[Server]
S -->|stdio| AGENT[AI Agent]
Shared mode — standalone broker + multiple MCP servers:
graph LR
US[Userscript] -->|HTTP :27342| BR[Bridge Broker]
MCP_A[MCP Server A] -->|/mcp/*| BR
MCP_B[MCP Server B] -->|/mcp/*| BR
AGENT_A[AI Agent A] -->|stdio| MCP_A
AGENT_B[AI Agent B] -->|stdio| MCP_B
iitc-mcp broker # start standalone broker
iitc-mcp serve --broker-url http://127.0.0.1:27342 # connect to shared broker
Commands are queued by ID — simultaneous operations may interfere. In practice, only one agent operates at a time.
Three packages:
packages/protocol— shared Zod schemaspackages/iitc-plugin— userscript (page adapter + transport)packages/mcp-server— Node.js MCP server (broker + HTTP + CLI)
Build & Development
git clone https://github.com/comicchang/iitc-mcp.git
cd iitc-mcp
npm ci --legacy-peer-deps
npm run build && npm test # 163 tests, typecheck, 3 build artifacts
Daily dev commands:
npm run typecheck # strict TypeScript
npm run build # userscript + server
npm run lint # ESLint
npm test # unit tests (163)
npm run test:smoke # no-browser smoke tests
# Start MCP server locally
npx tsx packages/mcp-server/src/cli.ts serve
# Read-only mode
npx tsx packages/mcp-server/src/cli.ts serve --read-only
License
See LICENSE. Fork must preserve the same license. Only Enlightened players may use this software. Resistance and Machina are not welcome. Attempting to bypass these restrictions is prohibited.
Enlightened 💚
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.