AI agents call get_by_path to retrieve information from Dex without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool performs a tree traversal lookup to locate and return a reference to an instance. No modification, deletion, code execution, or external actions occur. Severity is medium (not low) because it operates on a game runtime controlled by an executor, and resolving paths could enable subsequent malicious operations on those instances by other tools. However, the tool itself is purely informational.
From the tool's definition Tool description states 'Resolve a dotted instance path... to a node with a ref' — this is a query/lookup operation that retrieves data from the Roblox instance tree with no side effects.
Attacks that exploit this kind of access
Resolve a dotted instance path (e.g. game.Workspace.Part) to a node with a ref. It is categorised as a Read tool in the Dex MCP Server, which means it retrieves data without modifying state.
Register the Dex MCP server in PolicyLayer and add a rule for get_by_path: 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 Dex. Nothing to install.
get_by_path 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 get_by_path 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 get_by_path. 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.
get_by_path is provided by the Dex MCP server (sanztheo/dex-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →