mcp-debug

mcp-debug is a command for Claude Code from Matt-Dionis/claude-code-configs. It costs 8 tokens per session (698 once invoked), scanned B, original, MIT.

A command guide for diagnosing a Memory MCP server, a service that stores and searches information for coding assistants.

In plain words
What is it for?
Use it to debug server configuration, connection and protocol errors, memory CRUD tools, PostgreSQL vector operations, and embedding generation.
Why use it?
It gives a checklist for startup, JSON-RPC messages, tool registration, permissions, database access, vector search, and embedding problems.

Command for Claude Code

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/matt-dionis/claude-code-configs/mcp-debug
Clone the repo
git clone --depth 1 https://github.com/Matt-Dionis/claude-code-configs

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp-debug

README.md
[![agentmods](https://agentmods.dev/badge/commands/matt-dionis/claude-code-configs/mcp-debug.svg)](https://agentmods.dev/commands/matt-dionis/claude-code-configs/mcp-debug)
Your own site
<a href="https://agentmods.dev/commands/matt-dionis/claude-code-configs/mcp-debug"><img src="https://agentmods.dev/badge/commands/matt-dionis/claude-code-configs/mcp-debug.svg" alt="Measured on agentmods" height="20"></a>
Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 698 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.

ModelPer sessionOnce invoked
Fable 5 $0.00008 $0.00698
Opus 5 $0.00004 $0.00349
Sonnet 5 $0.00002 $0.00140
Haiku 4.5 $0.00001 $0.00070

Measured 4d ago against content hash a572c2c3cdfa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

mcp-debug 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 4d 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 MCP configurationmediumAgent snooping

mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.

cat .mcp.json
configurations/mcp-servers/memory-mcp-server/.claude/commands/mcp-debug.md · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory MCP Server Debugging

Debug the Memory MCP server implementation with focus on $ARGUMENTS:

1. Server Initialization & Configuration

  • Verify MCP server startup and registration
  • Check @modelcontextprotocol/sdk initialization
  • Validate server manifest and capabilities
  • Test stdio/HTTP transport configuration
  • Verify database connection (Neon PostgreSQL)

2. MCP Protocol Compliance

  • Validate JSON-RPC 2.0 message format
  • Test request/response correlation (id matching)
  • Verify error response format (code, message, data)
  • Check notification handling (no id field)
  • Validate batch request support

3. Memory Tool Registration

  • Verify tool discovery and registration:
    • create_memory - Memory creation with embeddings
    • search_memories - Vector similarity search
    • get_memory - Direct retrieval
    • update_memory - Memory updates
    • delete_memory - Soft/hard deletion
    • list_memories - Pagination support
  • Validate tool parameter schemas (Zod validation)
  • Test tool permission boundaries

4. Database & Vector Operations

  • Test pgvector extension functionality
  • Verify embedding generation (OpenAI API)
  • Debug vector similarity search queries
  • Check index usage (IVFFlat/HNSW)
  • Validate transaction handling

5. Session & Authentication

  • Debug companion session management
  • Verify user context isolation
  • Test multi-tenancy boundaries
  • Check session persistence
  • Validate auth token handling

6. Error Handling & Recovery

  • Test database connection failures
  • Handle embedding API errors
  • Verify graceful degradation
  • Check error logging and telemetry
  • Test retry mechanisms

7. Performance & Memory Leaks

  • Monitor connection pooling
  • Check for memory leaks in long sessions
  • Verify streaming response handling
  • Test concurrent request handling
  • Profile vector search performance

8. Common Issues & Solutions

Connection Refused

# Check if server is running
ps aux | grep "memory-mcp"
# Verify port binding
lsof -i :3000
# Test direct connection
npx @modelcontextprotocol/cli connect stdio "node ./dist/index.js"

Read the full file on GitHub · 116 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. 4d ago First seen · 116 lines · 8 tokens per session scan B a572c2c3cdfa

Subscribe to this mod's changes

mcp-debug is a command published in the GitHub repository Matt-Dionis/claude-code-configs (625 stars, last pushed 1y ago), licensed MIT. It adds 8 tokens to every session and 698 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it B with 1 finding (reads mcp configuration). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.