MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202611102354 A) filed by Tejvir on August 25, 2026, for Threat-Conditioned Mutation Of Agent Capability Credentials.
Inventor includes Tejvir.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: A computer-implemented control plane governs autonomous AI agents via task-scoped capability credentials driven by a machine-readable threat model. Threat entries define affected capability sets, deterministic detection signatures, and graduated response tiers, compiled into constant-time runtime detectors evaluated along a non- inferential enforcement path. An agent executes a multi-step task under a credential enumerating permitted operations, some single-use. On a detector match, a response engine mutates the credential by tier: contracting capabilities while preserving a minimum-safe floor, reissuing a shorter-lived credential scoped to the exercised subset, or quarantining it with a taint-scoped context purge. A fleet immuniser scans other agents' live credentials and pre-emptively contracts intersecting capabilities before detection occurs, shrinking fleet-wide attack surface. All issue, detection, mutation and completion events are recorded in a tamper-evident, hash-chained ledger anchored to a transparency log, and enforced by an attested enforcer.
Disclaimer: Curated by HT Syndication.
