mcp-console
Click on "Install 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., "@mcp-consoleList the files in the shared workspace"
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.
ctf-brain-user — participant client (MCP)
A stdio MCP server you connect to your CLI agent (Claude Code, Codex, Google Antigravity, or any MCP-capable agent) to work in your team's shared workspace.
Requirements
Node.js 22+
gitYour team's server URL and a personal API key (generate it on the Settings page).
Related MCP server: cursor-agent-bridge
Install
git clone https://github.com/sanjarbiy/ctf-brain-user
cd ctf-brain-user
npm install1. Add your keys (one-time)
On the Settings page, add at least 3 OpenRouter keys + 1 Jina key — or from the CLI:
node --import tsx client/src/cli.ts keys2. Generate your API key
On Settings → Connect API keys, generate a key (shown once — copy it). It carries your workspace, so BRAIN_CTF is not needed.
3. Connect the MCP to your agent
The MCP server is always the same command:
node --import tsx <ABSOLUTE-PATH>/ctf-brain-user/client/src/cli.ts mcpwith these env vars (use an absolute BRAIN_STATE_DIR — ~ is not expanded by every agent):
BRAIN_SERVER=https://your-team-server.example
BRAIN_TOKEN=mcp_your_generated_key
BRAIN_STATE_DIR=/home/you/.mcp-console # Windows: C:\Users\you\.mcp-console
BRAIN_VAULT_KEY=<64-hex> # openssl rand -hex 32Claude Code
claude mcp add mcp-console \
--env BRAIN_SERVER=https://your-team-server.example \
--env BRAIN_TOKEN=mcp_your_generated_key \
--env BRAIN_STATE_DIR=$HOME/.mcp-console \
--env BRAIN_VAULT_KEY=$(openssl rand -hex 32) \
-- node --import tsx /absolute/path/ctf-brain-user/client/src/cli.ts mcpRestart Claude Code.
Codex CLI
Option A — codex mcp add (everything after -- is the server command):
codex mcp add mcp-console \
--env BRAIN_SERVER=https://your-team-server.example \
--env BRAIN_TOKEN=mcp_your_generated_key \
--env BRAIN_STATE_DIR=/home/you/.mcp-console \
--env BRAIN_VAULT_KEY=<64-hex> \
-- node --import tsx /absolute/path/ctf-brain-user/client/src/cli.ts mcpOption B — edit ~/.codex/config.toml (Windows: C:\Users\you\.codex\config.toml):
[mcp_servers.mcp-console]
command = "node"
args = ["--import", "tsx", "/absolute/path/ctf-brain-user/client/src/cli.ts", "mcp"]
startup_timeout_sec = 60 # tsx compiles TS on first launch; raise from the default 10s
[mcp_servers.mcp-console.env]
BRAIN_SERVER = "https://your-team-server.example"
BRAIN_TOKEN = "mcp_your_generated_key"
BRAIN_STATE_DIR = "/home/you/.mcp-console"
BRAIN_VAULT_KEY = "<64-hex>"Set startup_timeout_sec = 60 even if you used Option A (edit the entry add created) — otherwise the first launch times out while tsx compiles. Restart Codex; verify with codex mcp list.
Google Antigravity ("agy")
Edit ~/.gemini/config/mcp_config.json (Windows: C:\Users\you\.gemini\config\mcp_config.json) — or in the IDE open … → MCP Servers → Manage MCP Servers → View raw config, or in the CLI type /mcp:
{
"mcpServers": {
"mcp-console": {
"command": "node",
"args": ["--import", "tsx", "/absolute/path/ctf-brain-user/client/src/cli.ts", "mcp"],
"env": {
"BRAIN_SERVER": "https://your-team-server.example",
"BRAIN_TOKEN": "mcp_your_generated_key",
"BRAIN_STATE_DIR": "/home/you/.mcp-console",
"BRAIN_VAULT_KEY": "<64-hex>"
}
}
}
}On Windows use forward slashes in the path (or escape backslashes). Save, then reload from the Manage MCP Servers panel (IDE) or /mcp (CLI).
hackerai
hackerai does not connect external MCP servers — it runs its own built-in tools in a sandbox. Two options to use this client from there:
Run it as a shell command in hackerai's terminal — install this repo on the sandbox host and call a CLI subcommand directly (see
cli.ts help), instead of the persistentmcpserver.Add it as a native tool by patching the hackerai source (advanced).
Safety
ctf_agent runs commands as local shell on your machine (guarded only against host-destructive commands like rm -rf /). Enable it only on a host where you accept local command execution. Make sure the client logs to stderr, not stdout — stdio MCP multiplexes JSON-RPC over stdout.
CLI
node --import tsx client/src/cli.ts helplogin / redeem (join with an invite code), keys (submit your OpenRouter/Jina keys), mcp (serve the MCP), status, doctor.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
No tool schema history has been recorded yet.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Connect AI agents to Filepad workspaces through OAuth MCP.
Connect AI agents to 1000+ apps with managed authentication and tool-calling.
Securely search and manage workspace context files for AI agents and teams.
The team layer for AI coding agents: shared contracts, collision alerts, E2EE sessions.
Related MCP Servers
- AlicenseAqualityCmaintenanceConnects Claude Code instances to a clawborrator hub via WebSocket, enabling cross-session communication, file exchange, and agent dispatch.12191MIT
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients like Claude Code to delegate coding tasks to the local Cursor Agent CLI, with persistent per-workspace sessions that resume across calls.12MIT
- AlicenseAqualityBmaintenanceEnables an AI agent to run local tools on the user's own machine via stdio, including command execution, workspace file read/write, and system status checks.52MIT
- AlicenseAqualityBmaintenanceConnects ChatGPT to a local developer workspace through MCP, enabling bounded repository analysis, file and image inspection, direct edits, command verification, and Git-aware review.22MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/sanjarbiy/ctf-brain-user'
If you have feedback or need assistance with the MCP directory API, please join our Discord server