AI agents call linktest_verify to retrieve information from Linktest without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool name 'linktest_verify' and the partial description suggest a verification or validation check (read-like operation) to confirm device access. The description appears truncated ('install the user' is incomplete), which lowers confidence. No write, execute, destructive, or financial actions are indicated by the available text. Severity is low as verification checks typically have minimal blast radius.
From the tool's definition 'Verify that LinkTest devices can actually access and install the user' — the description is incomplete/truncated but implies a verification/check operation
Attacks that exploit this kind of access
Verify that LinkTest devices can actually access and install the user. It is categorised as a Read tool in the Linktest MCP Server, which means it retrieves data without modifying state.
Register the Linktest MCP server in PolicyLayer and add a rule for linktest_verify: 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 Linktest. Nothing to install.
linktest_verify 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 linktest_verify 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 linktest_verify. 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.
linktest_verify is provided by the Linktest MCP server (linktest-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 →