AI agents call odx_config to retrieve information from Odxproxy without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a read-only operation that queries and returns existing configuration information. While the returned data (endpoint URLs, database names, user IDs) could be sensitive, the tool itself only retrieves data without side effects. The severity is low because accessing configuration information has limited direct impact compared to tools that can modify data or execute arbitrary operations.
From the tool's definition Tool retrieves and returns configuration data (server URL, database, gateway endpoint, user ID) without modifying or deleting any information.
Attacks that exploit this kind of access
Retrieve the ODXProxy client configuration, including server URL, database, gateway endpoint, and user ID. It is categorised as a Read tool in the Odxproxy MCP Server, which means it retrieves data without modifying state.
Register the Odxproxy MCP server in PolicyLayer and add a rule for odx_config: 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 Odxproxy. Nothing to install.
odx_config 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 odx_config 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 odx_config. 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.
odx_config is provided by the Odxproxy MCP server (terrakernel/odxproxy-mcpserver). 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 →