A setup wizard for connecting Gutt memory to Claude Code or Cursor, two coding environments. It detects the active environment and guides the MCP server connection.

Initialize Gutt integration, detect whether Claude Code or Cursor is running, check existing configuration, and complete the connection setup.

It avoids manual configuration mistakes and uses the server's own authorization flow instead of requiring users to construct login details.

setup — Command by iBrain-BVBA

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/ibrain-bvba/gutt-claude-code-plugin/setup

Clone the repo

git clone --depth 1 https://github.com/iBrain-BVBA/gutt-claude-code-plugin

Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 958 The whole file, excluding the scripts and references it only reads on demand.

Security scan B 1 finding. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00009	$0.00958
Opus 5	$0.00005	$0.00479
Sonnet 5	$0.00002	$0.00192
Haiku 4.5	$0.00001	$0.00096

Measured 2d ago against content hash 4dd32eb4fb25, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

setup scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- Do NOT read `~/.claude/settings.json` to find URLs

gutt-core/commands/setup.md · 140 lines

How it starts

The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.

gutt Setup Wizard

Configure gutt MCP server connection for Claude Code or Cursor.

CRITICAL RULES - READ FIRST

YOU MUST FOLLOW THESE RULES:

NO HARDCODED PATHS - Never use paths like C:\Users\username. Always use:
- Windows: %USERPROFILE% or $env:USERPROFILE (PowerShell)
- Unix: $HOME
NO MANUAL BROWSER OPENING - NEVER run Bash commands like:
- start "" "https://..."
- open https://...
- xdg-open https://...
NO CONSTRUCTED AUTH URLs - NEVER construct URLs like https://domain/auth/login. The MCP server returns auth URLs in its response if needed.
NO GROUP_ID QUESTION - The MCP server handles authorization automatically.

IDE Detection

Before starting, detect which IDE is running:

If environment variable CURSOR_PROJECT_DIR or CURSOR_VERSION is set → Cursor
Otherwise → Claude Code (also the backward-compatible default when nothing is set)

CURSOR_PLUGIN_ROOT is not a Cursor variable and must not be treated as a signal; CLAUDE_PLUGIN_ROOT is set by Claude Code but says nothing on its own, since Cursor leaves it unset rather than contradicting it.

Setup Flow

Step 1: Check Existing Configuration

Check if gutt-mcp-remote is already configured by attempting to list MCP resources.

If found, show the configured endpoint and ask: "Reconfigure?" / "Test connection?" / "Exit"

Step 2: Get MCP Endpoint URL

Ask the user directly for their URL (plain text question, NOT multiple choice):

Enter your organization's gutt MCP endpoint URL:
(Example: https://your-org.a.run.app/mcp)

Validate:

Must start with https://
Must end with /mcp

Step 3: Add MCP Server (IDE-specific)

Claude Code

Run the claude mcp add command:

claude mcp add gutt-mcp-remote --transport http --scope user "[USER_PROVIDED_URL]"

This registers the MCP server in Claude Code's user settings.

Cursor

Cursor has no mcp add CLI command. Register the MCP server by editing the Cursor MCP config file directly:

Read the full file on GitHub · 140 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 140 lines · 9 tokens per session scan B 4dd32eb4fb25

Subscribe to this mod's changes

setup is a command published in the GitHub repository iBrain-BVBA/gutt-claude-code-plugin (4 stars, last pushed 11d ago), licensed MIT. It adds 9 tokens to every session and 958 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Memory and context

setup