mcp-debugging

A set of instructions for diagnosing an MCP server that connects Prometheus services. MCP is a standard way for an AI agent to call tools provided by another service.

In plain words
What is it for?
Use it to list and inspect tools, call them for testing, check registered services, view unhealthy-service details, and restart a service when needed.
Why use it?
It provides a defined sequence for checking service health, server registration, available tools, and connection problems.

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

Made for: Cursor.

Per session 486 This file is loaded in full into every session.
When invoked 486 The same file — it is already loaded in full.
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 $0.00486 $0.00486
Opus 5 $0.00243 $0.00243
Sonnet 5 $0.00097 $0.00097
Haiku 4.5 $0.00049 $0.00049

Measured yesterday against content hash 438fcf3ca448, 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 · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debugging mcp-prometheus via mcp-debug

mcp-debug provides core debugging tools to test and explore the functionality implemented in the aggregated MCP server. This guide focuses on mcp-debugs internal debugging capabilities.

Core Debugging Tools

mcp-debug exposes these categories of internal tools for debugging the aggregated mcp server end-to-end:

list_tools: list all the tools exposed by the aggregated mcp server describe_tool: describe a tool exposed by the aggregated mcp server call_tool: execute a tool exposed by the aggregated mcp server

Debugging Workflow

1. Verify All Services Are Running

Start by checking the overall health of the system:

mcp_mcp-debug_call_tool(name="core_service_list", arguments={})

2. Check MCP Server Registration

Verify that all expected MCP servers are properly registered with the aggregator:

mcp_mcp-debug_call_tool(name="core_mcp_server_list", arguments={})

5. Troubleshoot Connection Issues

If a service shows as unhealthy:

  1. Check its detailed status: core_service_status
  2. Try restarting it: core_service_restart
  3. Check if its MCP client is attached in the service list

Important Notes

  • The prefix x_ is configurable via EnvctlPrefix in the aggregator config
  • Tool names from individual MCP servers get prefixed to avoid conflicts
  • The agent automatically handles tool name resolution and routing
  • Service health is continuously monitored and reflected in the status
  • Workflows provide a way to test complex multi-step operations

Common Issues and Solutions

"CallTool not implemented" errors

  • This means the MCP client is not properly attached to the service
  • Check core_service_list to see if the service shows a client
  • The fix involves ensuring GetServiceData() returns the MCP client

Service shows as unhealthy

  • Check logs in the mcp-prometheus TUI (if available)
  • Use core_service_status for detailed information
  • Try core_service_restart to recover

Workflow validation fails

  • Use core_workflow_validate to check syntax
  • Ensure tool names exist (check with mcp_mcp-debug_list_tools)
  • Verify argument schemas match the tool requirements

Read the full file on GitHub · 62 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 · 62 lines · 486 tokens per session scan A 438fcf3ca448

Subscribe to this mod's changes

mcp-debugging is a cursor rule published in the GitHub repository giantswarm/mcp-prometheus (10 stars, last pushed 2d ago), licensed Apache-2.0. It adds 486 tokens to every session, about $0.0024 per session on Opus 5. 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 2026-08-31.