AI agents call get_trace_parameter to retrieve information from Openl without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a read-only query to retrieve a parameter value from an already-traced execution context. It does not create, modify, delete, or execute anything—it simply fetches data. The lazy-loading mechanism is an implementation detail of data retrieval, not an operation that modifies state or triggers external effects.
From the tool's definition The tool description states 'Get lazy-loaded parameter value,' which is a retrieval operation with no side effects. It fetches a parameter value that has already been loaded or is available for lazy loading.
Attacks that exploit this kind of access
Get lazy-loaded parameter value. Use when a TraceParameterValue has lazy:true and parameterId set. It is categorised as a Read tool in the Openl MCP Server, which means it retrieves data without modifying state.
Register the Openl MCP server in PolicyLayer and add a rule for get_trace_parameter: 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 Openl. Nothing to install.
get_trace_parameter 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_trace_parameter 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_trace_parameter. 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_trace_parameter is provided by the Openl MCP server (openl-mcp-server). 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 →