todo_read · MCP Claude Code MCP: Risk & Policy

WHAT IT DOES

What todo_read does on MCP Claude Code

AI agents call todo_read to retrieve information from MCP Claude Code without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

WHY IT NEEDS A POLICY

Why todo_read needs a policy

The naming convention 'read' suffix and context among modify-capable siblings indicate this retrieves data without modification. However, confidence is moderate (0.72) because the description is empty and the exact scope of TODO data is not specified. If it merely lists TODO items from a codebase or project file, it is a simple Read operation with low severity (minimal blast radius if misused).

From the tool's definition Tool name 'todo_read' strongly suggests a read-only operation ('read' suffix). Given the server context of file system interactions and the sibling tools including 'content_replace', 'edit', and 'multi_edit' (clearly write/destructive), the 'todo_read'…

Attacks that exploit this kind of access

Questions about todo_read

What does the todo_read tool do? +

todo_read. It is categorised as a Read tool in the MCP Claude Code MCP Server, which means it retrieves data without modifying state.

How do I enforce a policy on todo_read? +

Register the MCP Claude Code MCP server in PolicyLayer and add a rule for todo_read: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches MCP Claude Code. Nothing to install.

What risk level is todo_read? +

todo_read is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit todo_read? +

Yes. Add a rate_limit block to the todo_read rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block todo_read completely? +

Set action: deny in the PolicyLayer policy for todo_read. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides todo_read? +

todo_read is provided by the MCP Claude Code MCP server (sdglbl/mcp-claude-code). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.