kubernetes_proxy
AI agents invoke kubernetes_proxy to trigger actions in Portainer MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
A Kubernetes proxy tool typically forwards API requests to a Kubernetes cluster, enabling operations ranging from read-only queries to destructive actions like deleting pods or namespaces. Given the sibling tool 'docker_proxy' (which similarly proxies Docker operations), this tool likely acts as a general-purpose proxy to the Kubernetes API.
From the tool's definition Tool name 'kubernetes_proxy' implies proxying requests to a Kubernetes API, which could execute arbitrary Kubernetes operations. Description is empty, lowering confidence.
Attacks that exploit this kind of access
kubernetes_proxy. It is categorised as a Execute tool in the Portainer MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Portainer MCP server in PolicyLayer and add a rule for kubernetes_proxy: 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 Portainer MCP. Nothing to install.
kubernetes_proxy is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the kubernetes_proxy 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 kubernetes_proxy. 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.
kubernetes_proxy is provided by the Portainer MCP server (portainer/portainer-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 →