tool_get_sample
AI agents call tool_get_sample to retrieve information from Inspect Logs without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Despite the empty description, context strongly suggests this tool retrieves sample data from evaluation logs. The naming pattern and sibling tools (all Read operations) indicate it queries or fetches specific log samples without side effects. Confidence is moderate due to missing description, but the server's stated function and peer tools support Read classification.
From the tool's definition Tool name 'tool_get_sample' with no description; sibling tools are read-only operations (list_logs, search_logs, get_eval_summary, get_aggregate_stats, compare_runs). Server purpose is to 'explore and analyze' logs, indicating inspection without modification.
Attacks that exploit this kind of access
tool_get_sample. It is categorised as a Read tool in the Inspect Logs MCP Server, which means it retrieves data without modifying state.
Register the Inspect Logs MCP server in PolicyLayer and add a rule for tool_get_sample: 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 Inspect Logs. Nothing to install.
tool_get_sample 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 tool_get_sample 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 tool_get_sample. 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.
tool_get_sample is provided by the Inspect Logs MCP server (pranshusrivastava/inspect-logs-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 →