get_pod_logs
AI agents call get_pod_logs to retrieve information from Loki MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves log data from Kubernetes pods via Grafana Loki. The description is empty, but context from sibling tools and server purpose confirms this is a query/retrieval operation with no ability to modify, delete, or execute arbitrary actions. Reading logs has minimal blast radius unless the logs themselves contain sensitive data, but the operation itself is non-destructive and side-effect-free.
From the tool's definition Tool name 'get_pod_logs' and sibling tools ('search_logs', 'find_pod_restarts', 'get_error_summary') indicate read-only log retrieval operations.
Attacks that exploit this kind of access
get_pod_logs. It is categorised as a Read tool in the Loki MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Loki MCP Server MCP server in PolicyLayer and add a rule for get_pod_logs: 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 Loki MCP Server. Nothing to install.
get_pod_logs 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_pod_logs 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_pod_logs. 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_pod_logs is provided by the Loki MCP Server MCP server (mvs5465-test/loki-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 →