// MCP TOOL REFERENCE

SHELBYMCP TOOLS

9 tools from the Shelbymcp MCP Server, categorised by risk level.

READ TOOLS

capture_thought Store a thought with metadata, topics, and relationships explore_graph Traverse the knowledge graph from a starting thought get_thought Retrieve a specific thought by ID with full content list_thoughts Browse and filter thoughts by type, topic, person, project, or date range search_thoughts Full-text search with optional vector reranking. Returns summaries, not full content. 2/5 thought_stats Aggregate statistics about the memory database

WRITE TOOLS

manage_edges Create or remove typed relationships between thoughts 2/5 update_thought Update content or metadata on one or more thoughts 3/5

DESTRUCTIVE TOOLS

delete_thought Remove a thought and its edges 4/5

// FAQ

How many tools does the Shelbymcp MCP server have? +

The Shelbymcp MCP server exposes 9 tools across 3 categories: Read, Write, Destructive.

How do I enforce policies on Shelbymcp tools? +

Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Shelbymcp server.

What risk categories do Shelbymcp tools fall into? +

Shelbymcp tools are categorised as Read (6), Write (2), Destructive (1). Each category has a recommended default policy.

Enforce policies on Shelbymcp

Open source. One binary. Zero dependencies.

npx -y @policylayer/intercept

github.com/policylayer/intercept →