AI agents call search_todos to retrieve information from Things 3 MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries existing task data based on search criteria. It has no side effects—it does not create, modify, delete, or execute any actions. The blast radius of misuse is minimal; an AI agent could only over-retrieve task information, which poses no risk to data integrity or system state.
From the tool's definition Tool name 'search_todos' and description 'Search todos by title or notes' indicate a read-only operation that retrieves data without modification.
Documented attack patterns abuse exactly the kind of access search_todos gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Things 3 MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for search_todos:
{
"version": "1",
"default": "deny",
"tools": {
"search_todos": {}
}
}search_todos is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Search todos by title or notes. It is categorised as a Read tool in the Things 3 MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Things 3 MCP Server MCP server in PolicyLayer and add a rule for search_todos: 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 Things 3 MCP Server. Nothing to install.
search_todos is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the search_todos 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.
Set action: deny in the PolicyLayer policy for search_todos. 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.
search_todos is provided by the Things 3 MCP Server MCP server (rossshannon/things3-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 24 Things 3 MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
24 Things 3 MCP Server tools catalogued and risk-classified — across an index of 42,500+ MCP servers.