Abstract

AI agents rely on third-party tools, such as Model Context Protocol (MCP) servers, whose publishers can be authenticated and whose execution can be restricted by runtime security policies. Yet authenticity does not guarantee benign behaviour: a legitimately signed tool may perform its advertised function while concealing malicious code that leaks confidential information. We demonstrate this trust failure with a signed MCP tool that embeds sensitive data in requests to a policyallowed destination. Metas MCPGuard-Dynamic, a state-of-theart runtime defence, blocks a direct exfiltration attempt in the same run, yet fails to prevent this attack in every trial. We introduce germline execution, which limits information disclosure without inspecting or trusting tool implementations. Each tool invocation runs in ephemeral copies forked from a datafree germline process, each receiving only information authorised for its destination. Even if malicious code executes, what it can disclose is bounded by the data available to its copy. Under explicit assumptions on copy confinement and a datafree germline, we prove sink noninterference for every signed release and, for releases deterministic within our replay model, show that a verified witness establishes a flow violation. Across 153 attack runs, no unauthorised disclosure was detected under germline execution, compared with 5.974.5% of runs under baseline defences and an ablation. Invocations served by a single copy completed in 1049 ms, 228% faster than restarting the tool for every invocation, the conventional way to clear state. These results show that data-isolated execution protects agent toolchains against malicious code concealed within authenticated software.