AI agents call snapshot to retrieve information from Wraith without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The snapshot tool queries the DOM to extract and return information about page elements. While it operates within a browser automation context (where the underlying server has bot-detection bypass capabilities), the tool itself performs only a read operation — it captures and returns element data without creating, modifying, deleting, or executing any actions.
From the tool's definition Tool description states 'Return the current page's interactive elements' — this is a retrieval operation that reads and returns data about the current page state with no modification or side effects.
Attacks that exploit this kind of access
Return the current page's interactive elements as [index] <tag> text. It is categorised as a Read tool in the Wraith MCP Server, which means it retrieves data without modifying state.
Register the Wraith MCP server in PolicyLayer and add a rule for snapshot: 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 Wraith. Nothing to install.
snapshot 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 snapshot 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 snapshot. 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.
snapshot is provided by the Wraith MCP server (koreahwan/wraith-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 →