Metraipolis › archive › dc_df003b694871
Signed but Not Safe: Requirement Integrity in Multi-Agent LLM Handoffs — arXiv:2611.07483
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.07483
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-06T14:41:26.628Z
- SHA-256 of the archived text
cbf6945756a755b83a6e14c2903d6ed6d350eed7fcc731e2daae105191bf21e3- Stable URL of this page
https://metraipolis.com/documento/dc_df003b694871
How to cite
Signed but Not Safe: Requirement Integrity in Multi-Agent LLM Handoffs — arXiv:2611.07483 — arXiv (en), https://arxiv.org/abs/2611.07483 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_df003b694871 (Metraipolis archive dc_df003b694871, sha256 cbf6945756a755b83a6e14c2903d6ed6d350eed7fcc731e2daae105191bf21e3)
BibTeX
@misc{dcdf003b694871,
title = {Signed but Not Safe: Requirement Integrity in Multi-Agent LLM Handoffs — arXiv:2611.07483},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.07483},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_df003b694871; sha256 cbf6945756a755b83a6e14c2903d6ed6d350eed7fcc731e2daae105191bf21e3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07483. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Enterprise AI agents increasingly split a user's request across subagents and merge their results, and each handoff can summarize away, alter or ignore a requirement. We report a preregistered study with synthetic, exactly scorable tasks comparing free-text, structured JSON and signed requirement-manifest handoffs: the orchestrator extracts each requirement as a verbatim quote plus a checkable predicate, code signs the manifest with HMAC-SHA256, and deterministic code verifies the merger's decision. With Qwen3-8B (1,440 trajectories, five threats), signed manifests cut unsafe commits under attack from 69.4% to 9.0% (60.4 pp, 95% CI [51.4, 69.4]) with no net loss of conformant commits relative to structured handoffs. Signing is what defeats in-transit tampering: an unsigned manifest committed violations in 79% of mutated runs, the signed one in none. On identical trajectories the deterministic verifier caught 78.7% of violating decisions versus 59.0% for an LLM checker, with 7% versus 50% false alarms. An oracle manifest removed all remaining unsafe commits: the residual risk comes from extraction errors made before signing. A preregistered pre-signing injection attack confirmed this boundary: one adversarial line raised signed-manifest violations from 14.6% to 37.5% (Qwen3-8B), 11.1% to 63.9% (gpt-oss-20b) and 8.3% to 75.0% (GPT-5.6-luna), and the signature then authenticated the corrupted manifest. Signing protects requirements in transit, not at extraction; the extraction step is the trust boundary. 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/2611.07483 — 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 2026-11-06T14:18:27.668Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).