mcp-debugging

A debugging guide and command-line agent for MCP, the protocol that lets AI assistants use tools and resources from other services.

In plain words
What is it for?
Use it to inspect an MCP server, interactively run its capabilities with JSON arguments, or expose debugging tools to an AI assistant.
Why use it?
It makes an MCP server's available tools, resources, prompts, notifications, and responses visible and testable.

Cursor rule for Cursor

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 rules/giantswarm/mcp-debug/mcp-debugging
Clone the repo
git clone --depth 1 https://github.com/giantswarm/mcp-debug

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 1,136 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01136
Opus 5 $0.00000 $0.00568
Sonnet 5 $0.00000 $0.00227
Haiku 4.5 $0.00000 $0.00114

Measured yesterday against content hash 67365c333b2f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp-debugging scanned grade A with 0 findings 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 yesterday.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/rules/mcp-debugging.mdc · 180 lines

How it starts

The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debugging MCP Servers via MCP-Debug Agent

The mcp-debug agent provides essential tools to test and explore MCP (Model Context Protocol) servers. This guide focuses on using the agent to debug MCP server behavior and functionality.

Core Agent Modes

The agent operates in three distinct modes:

1. Normal Mode (Default)

Connects to an MCP server, lists available tools/resources/prompts, and waits for notifications:

mcp-debug --endpoint http://localhost:8090/sse

2. REPL Mode

Interactive mode for exploring and executing MCP capabilities:

mcp-debug --repl

In REPL mode, you can:

  • List available tools, resources, and prompts
  • Get detailed information about specific items
  • Execute tools interactively with JSON arguments
  • View resources and retrieve their contents
  • Execute prompts with arguments
  • Toggle notification display

3. MCP Server Mode

Run the agent as an MCP server exposing debugging tools via stdio:

mcp-debug --mcp-server

This mode is designed for integration with AI assistants like Claude or Cursor. Configure it in your AI assistant's MCP settings.

REPL Commands

Listing Commands

list tools                   # List all available tools
list resources               # List all available resources
list prompts                 # List all available prompts

Describe Commands

describe tool <name>         # Show detailed information about a tool
describe resource <uri>      # Show detailed information about a resource
describe prompt <name>       # Show detailed information about a prompt

Execution Commands

call <tool> {json}           # Execute a tool with JSON arguments
get <resource-uri>           # Retrieve a resource
prompt <name> {json}         # Get a prompt with JSON arguments

Control Commands

notifications <on|off>       # Enable/disable notification display
help, ?                      # Show help message
exit, quit                   # Exit the REPL

Read the full file on GitHub · 180 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.

  1. yesterday First seen · 180 lines · 0 tokens per session scan A 67365c333b2f

Subscribe to this mod's changes

mcp-debugging is a cursor rule published in the GitHub repository giantswarm/mcp-debug (37 stars, last pushed 5d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,136 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.