NetGrind MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@NetGrind MCP Server List the submissions waiting in my Judgement Hall."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
NetGrind MCP server
Connect NetGrind to ChatGPT, Claude and any AI agent through Cotlet.dev. Work in your own Game Master account with 3 hosted tools for the Judgement Hall, Player progress and Shop rewards. Both write tools ask for approval by default.
What you can ask
List the submissions waiting in my Judgement Hall.
Show which Players are stuck or quiet.
List the rewards in my Shop and their Gold prices.
Related MCP server: awesome-ai-native-cloudflare-app MCP Server
Connect in 2 minutes
Get your API key at https://cotlet.dev/app and export it as COTLET_API_KEY.
ChatGPT
Settings → Apps & Connectors → Create, with connector URL https://cotlet.dev/mcp/v1/connect, then sign in with Cotlet.dev. No API key needed.
Claude
Settings → Connectors → Add custom connector, with URL https://cotlet.dev/mcp/v1/connect, then sign in with Cotlet.dev.
Claude Code
claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1 --header "Authorization: Bearer $COTLET_API_KEY"Cursor
Copy .cursor/mcp.json from this repo into your project:
{
"mcpServers": {
"cotlet": {
"url": "https://cotlet.dev/mcp/v1",
"headers": {
"Authorization": "Bearer ${COTLET_API_KEY}"
}
}
}
}Run it locally (stdio)
This repo is a real MCP server. It speaks MCP over stdio, lists its NetGrind tools without any key, and forwards every tool call to the hosted Cotlet.dev server. The hosted URL above stays the recommended way. Use this when your client only starts local servers.
Run the included Python module with Python 3.10 or later:
python3 netgrind_agent.py serveOr with Docker:
docker build -t netgrind-mcp .
docker run -i --rm -e COTLET_API_KEY netgrind-mcpTools: netgrind_guide, netgrind_status, netgrind_open, netgrind_read, netgrind_click, netgrind_type, netgrind_close. Without COTLET_API_KEY a tool call answers: "Set COTLET_API_KEY. Get a key at https://cotlet.dev/app".
Command line
Python 3.10+, with no third-party runtime dependencies. Clone this repository and run:
export COTLET_API_KEY="your-key"
python3 netgrind_agent.py connect # ready config for Claude Code, Cursor, ChatGPT
python3 netgrind_agent.py status # is NetGrind ready in your Cotlet.dev account
python3 netgrind_agent.py guide # NetGrind screens and sign-in link
python3 netgrind_agent.py read # open NetGrind, read the page, close the browser
python3 netgrind_agent.py tools # list Cotlet.dev tools
python3 netgrind_agent.py serve # start the stdio MCP server (same as no argument)Questions from NetGrind users
Which NetGrind account should I connect?
Sign in at https://play.netgrind.online/login in your Cotlet Cloud Browser with your Game Master account. Complete Discord, Google or email sign-in yourself.
Which actions ask for approval?
review_submissions and manage_shop_rewards ask for approval by default. track_player_progress only reads.
Is Cotlet affiliated with NetGrind?
No. Cotlet.dev is independent and not affiliated with NetGrind.
Learn more
Request access: https://cotlet.dev/signup?app=netgrind&utm_source=github&utm_medium=readme&utm_campaign=netgrind
Workflows on the hosted integration
The live integration page documents these 3 operations:
review_submissionstrack_player_progressmanage_shop_rewards
The local stdio adapter exposes seven browser helpers listed above. These helpers forward to Cotlet.dev; the workflow operations are documented on the hosted integration page.
Available Tools
7 toolsnetgrind_clickB
Click a numbered element on the current NetGrind page.
Inputs: element (required).
netgrind_closeA
Close the NetGrind browser session when the work is done, so credits stop running.
Inputs: no named parameters.
netgrind_guideA
How to work in NetGrind (review submissions, track Players and manage Shop rewards): its screens and the sign-in link. Call this first.
Inputs: task.
netgrind_openA
Open a NetGrind page in your signed-in Cloud Browser (default https://play.netgrind.online) to work with review submissions, track Players and manage Shop rewards.
Inputs: url.
netgrind_readA
Read the current NetGrind page: its text and numbered buttons, links and fields.
Inputs: element.
netgrind_statusA
Check that your Cotlet.dev account and NetGrind are ready: account, credits and NetGrind connection.
Inputs: no named parameters.
netgrind_typeB
Type text into a numbered field on the current NetGrind page.
Inputs: text (required), element (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Each tool maps to a clearly distinct step in a browser workflow: onboarding (guide), readiness check (status), session lifecycle (open/close), and page interaction (read/click/type). There is no meaningful overlap — read only inspects, click/type only act on numbered elements, and guide/status are explicitly differentiated by their wording.
All seven tools share a uniform netgrind_ snake_case prefix, which makes the set highly predictable. The only minor deviation is that guide and status are nouns while open/read/click/type/close are verbs, though this is idiomatic and not confusing.
Seven tools is well-scoped for a single-site browser automation server: one discovery step, one session-setup step, three interaction primitives, one lifecycle teardown, and one status check. Every tool earns its place with no filler.
The core loop — open a page, read its elements, click and type, then close — plus onboarding and readiness checks covers the stated review/Player/Shop workflows. Minor gaps exist for navigation (back/forward), scrolling on long pages, and explicit submit/select actions, which agents must work around via numbered buttons and links.
Maintenance
Related MCP Connectors
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
Run your Krtuworks apps from an AI agent: setup, remote config, stats, players, bug reports.
Let your AI agents hand work to each other, share files, and use apps you approve.
OAuth 2.1 short-link tools for AI agents with scoped tokens, approvals, audit logs, and revocation.
Related MCP Servers
- AlicenseBqualityBmaintenanceDiscord server management MCP for AI agents: chat, moderation, admin, and full server building. 160+ tools, confirmations the model cannot answer, undo. Formerly Omnicord.164530 npm12Apache 2.0
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to interact with the app via MCP tools and resources, supporting self-improvement through goal-setting and metrics tracking on Cloudflare Workers.5MIT
- AlicenseAqualityBmaintenanceEnables an AI agent to act as a DM/KP in a CozyVTT campaign by narrating over chat, rolling server-authoritative dice, moving tokens, switching maps, managing initiative, and accessing character sheets through MCP.411MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to run a SaaS app's admin through a scoped, audited API — looking up users, applying support actions, reading growth and retention, building segments, toggling feature flags and drafting emails in plain language. Access is limited by key scopes, so tools outside a key's permissions never appear.MIT