Metraipolis › archive › dc_c5301404aea7

Signed, Sealed, Leaked: Trust Beyond Signature for MCP Toolchains — arXiv:2611.06585

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.06585
License (as archived)
CC0-1.0
Archived
2026-11-05T18:16:25.717Z
SHA-256 of the archived text
2dd8f60704df3fbe969a17c065ff32151d53e88b584daae70752c54092f07c3e
Stable URL of this page
https://metraipolis.com/documento/dc_c5301404aea7

How to cite

Signed, Sealed, Leaked: Trust Beyond Signature for MCP Toolchains — arXiv:2611.06585 — arXiv (en), https://arxiv.org/abs/2611.06585 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c5301404aea7 (Metraipolis archive dc_c5301404aea7, sha256 2dd8f60704df3fbe969a17c065ff32151d53e88b584daae70752c54092f07c3e)
BibTeX
@misc{dcc5301404aea7,
  title = {Signed, Sealed, Leaked: Trust Beyond Signature for MCP Toolchains — arXiv:2611.06585},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.06585},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c5301404aea7; sha256 2dd8f60704df3fbe969a17c065ff32151d53e88b584daae70752c54092f07c3e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06585. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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 policy-allowed destination. Meta's MCPGuard-Dynamic, a state-of-the-art 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 data-free 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 data-free 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 5.9-74.5% of runs under baseline defences and an ablation. Invocations served by a single copy completed in 10-49 ms, 2-28x 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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2611.06585 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-05T18:10:09.311Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).