Zero-Trust Runtime Verification for Agentic Payment Protocols: Mitigating Replay and Context-Binding Failures in AP2
Qianlong Lan, Anuj Kaul, Shaun Jones, Stephanie Westrum · 2026
Read Zero-Trust Runtime Verification for Agentic Payment Protocols: Mitigating Replay and Context-Binding Failures in AP2 by Qianlong Lan, Anuj Kaul, Shaun Jones, Stephanie Westrum (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-06345.txt.
Description
AI agents that can pay for things act on signed mandates rather than live human approval, under protocols like the Agent Payments Protocol. Qianlong Lan and 3 co-authors audit how those mandates behave at run time. Retries, parallel calls, and orchestration layers strain the assumptions baked into a mandate when it was issued.
They propose a zero-trust verification layer that checks each use as it happens. Fresh, short-lived nonces bind a mandate to its exact context, and each mandate can be consumed only once. Simulations under heavy concurrency showed the two guards stop different failure classes: replay attacks and context mix-ups. The work argues authorization should be decided at execution time, not presumed from issuance.
Catalog record
| CREATOR | Qianlong Lan, Anuj Kaul, Shaun Jones, Stephanie Westrum |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-06345