diff_checkpoints
AI agents call diff_checkpoints to retrieve information from Kustomize MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Based on the name and server context, diff_checkpoints likely compares two saved checkpoints to show configuration differences — a read/query operation with no side effects. The empty description lowers confidence. Severity is medium because misuse could expose sensitive Kubernetes configuration details across environments.
From the tool's definition Tool name 'diff_checkpoints' and server context mentions 'comparing configuration changes through a checkpointing system' and 'manifest diffs across environments'
Attacks that exploit this kind of access
diff_checkpoints. It is categorised as a Read tool in the Kustomize MCP MCP Server, which means it retrieves data without modifying state.
Register the Kustomize MCP server in PolicyLayer and add a rule for diff_checkpoints: 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 Kustomize MCP. Nothing to install.
diff_checkpoints 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 diff_checkpoints 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 diff_checkpoints. 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.
diff_checkpoints is provided by the Kustomize MCP server (mbrt/kustomize-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 →