InfoResearchPreprint
Tether: Verified Stateful Policy Engine for System-Level Monitoring of Agents
- Published
- Record updated
Summary
Tether is a framework that enforces stateful security policies on the observed side effects of an AI agent's shell or program actions. Each proposed action runs in an ephemeral, isolated sandbox, and the policies, written in an embedded Lean 4 DSL, can be formally verified against global invariants. Evaluated on coding-agent benchmarks, a single shared policy blocked 100% of in-scope attacks with a benign false-positive rate under 5%.
Related items
- MediumCVE-2026-54026: LibreChat message injection via conversationId in request bodySimilar attack · NVD/CVE Database
- MediumCVE-2026-54046: LibreChat unauthenticated account enumeration via /api/user/verifySimilar attack · NVD/CVE Database
- HighCVE-2026-54044: LibreChat credential exposure through agent action domain repointingSimilar attack · NVD/CVE Database
- MediumCVE-2026-54043: LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.7…Similar attack · NVD/CVE Database
- MediumCVE-2026-54042: LibreChat 2FA verification unlimited code guessing via temp token endpointSimilar attack · NVD/CVE Database