MCP App Store
by KeenEthics
Rollcage icon

Rollcage MCP Server for Claude and ChatGPT

By Nicolae Maties

Overview

Rollcage gives Claude a safety check to run before it acts. Before it runs a shell command, opens a file or sends data to a site, Claude can ask Rollcage whether the action would be blocked, needs your approval, or is fine, and why. Rollcage uses the rules of rollcage-guard, an open-source project (MIT licence), with their default settings. The server is free and public, needs no account or API key, and stores nothing. It runs on rollcage.app, so it can’t see or change anything on your machine. Tools check_command: takes a shell command and returns block, ask or allow, with the reason. It never runs the command. Blocks: reading secret files commands that print environment variables (env, printenv, echo $API_KEY) one-line scripts that read environment variables (python -c, node -e…) rm -rf ~, git clean -f, DROP DATABASE, erasing disks uploads to paste, file-drop and tunnel sites Asks first: git reset --hard, force pushes, pushes to main or production curl | sh, sudo, terraform destroy deleting cloud resources, DELETE requests to remote APIs connecting to remote databases check_file: checks a list of paths and says which are secret files an agent shouldn’t read: .env files, keys and certificates, cloud credentials, service-account JSON, Terraform state, ~/.ssh, ~/.aws. Templates such as .env.example are allowed. check_url: checks a URL or host against the list of paste, file-drop, request-capture and tunnel sites (pastebin, transfer.sh, webhook.site, ngrok, trycloudflare…), which are the usual way keys and code leave a machine. It can also apply a stricter mode that only allows listed sites. list_rules: the full default rule set, with an action for each rule. Claude uses it to explain why something was flagged. install_guard: install steps for the local rollcage-guard hook, which enforces the same rules on every tool call in Claude Code, Codex CLI, Cursor and Gemini CLI, plus the hook entry for each agent’s config file. search_docs / get_doc: Rollcage’s guides to ru

How to connect

  1. Claude: Open the Rollcage listing in the Claude connectors directory and choose Connect. Claude listing Rollcage exposes 8 tools here.
  2. ChatGPT: Open the Rollcage listing in the ChatGPT app directory and choose Connect. ChatGPT listing Rollcage exposes 8 tools here.

Tools

check_command

ChatGPT
Runs a shell command through the Rollcage Guard's rules without running it. Returns block (secret files, environment dumps, inline code reading env vars, rm -rf ~, git clean -f, DROP DATABASE, disk erase, exfiltration sites), ask (git reset --hard, force push, curl | sh, sudo, terraform destroy, cloud deletes, DELETE requests, remote databases, pushes to protected branches) or allow, with the reason.

check_file

ChatGPT
Whether the Rollcage Guard treats each path as a secret file an agent must not read or write: .env and .env.* (but not .env.example), .envrc, .npmrc, .netrc, keys and certificates, service-account JSON, Rails master.key, Terraform state and tfvars, ~/.ssh, ~/.aws, ~/.kube/config and more.

check_url

ChatGPT
Whether the Rollcage Guard lets agents contact a URL or host. Paste, file-drop, request-capture and tunnel sites (pastebin.com, transfer.sh, webhook.site, ngrok, trycloudflare.com…) are blocked, because they are the usual way secrets and code leave a machine. Local and private addresses are always allowed.

get_doc

ChatGPT
One Rollcage feature page or guide as Markdown, by slug (from search_docs).

install_guard

ChatGPT
Steps to enforce the Rollcage Guard on every tool call, with the open-source rollcage-guard hook (macOS and Linux) or the Rollcage Mac app. Includes the hook entry for each agent's config file.

list_rules

ChatGPT
The Rollcage Guard's default rules: secret-file patterns, environment checks, command rules with their actions, blocked and allowed sites, and protected branches.

rollcage_info

ChatGPT
What Rollcage is, the latest version and what changed, price, system requirements and links.

search_docs

ChatGPT
Searches Rollcage's feature pages and its guides to running AI coding agents safely (Claude Code hooks, permissions, sandboxing, YOLO mode, undo, usage and cost, Codex, Cursor security…). Returns slugs for get_doc.

check_command

Claude
Runs a shell command through the Rollcage Guard's rules without running it. Returns block (secret files, environment dumps, inline code reading env vars, rm -rf ~, git clean -f, DROP DATABASE, disk erase, exfiltration sites), ask (git reset --hard, force push, curl | sh, sudo, terraform destroy, cloud deletes, DELETE requests, remote databases, pushes to protected branches) or allow, with the reason.

check_file

Claude
Whether the Rollcage Guard treats each path as a secret file an agent must not read or write: .env and .env.* (but not .env.example), .envrc, .npmrc, .netrc, keys and certificates, service-account JSON, Rails master.key, Terraform state and tfvars, ~/.ssh, ~/.aws, ~/.kube/config and more.

check_url

Claude
Whether the Rollcage Guard lets agents contact a URL or host. Paste, file-drop, request-capture and tunnel sites (pastebin.com, transfer.sh, webhook.site, ngrok, trycloudflare.com…) are blocked, because they are the usual way secrets and code leave a machine. Local and private addresses are always allowed.

get_doc

Claude
One Rollcage feature page or guide as Markdown, by slug (from search_docs).

install_guard

Claude
Steps to enforce the Rollcage Guard on every tool call, with the open-source rollcage-guard hook (macOS and Linux) or the Rollcage Mac app. Includes the hook entry for each agent's config file.

list_rules

Claude
The Rollcage Guard's default rules: secret-file patterns, environment checks, command rules with their actions, blocked and allowed sites, and protected branches.

rollcage_info

Claude
What Rollcage is, the latest version and what changed, price, system requirements and links.

search_docs

Claude
Searches Rollcage's feature pages and its guides to running AI coding agents safely (Claude Code hooks, permissions, sandboxing, YOLO mode, undo, usage and cost, Codex, Cursor security…). Returns slugs for get_doc.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/rollcage"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/rollcage)