collaborators

5 papers

cs.AI2026

Qualixar OS: A Universal Operating System for AI Agent Orchestration

Varun Pratap Bhardwaj

We present Qualixar OS, the first application-layer operating system for universal AI agent orchestration. Unlike kernel-level approaches (AIOS) or single-framework tools (AutoGen,…

cs.AI2026

AgentAssay: Token-Efficient Regression Testing for Non-Deterministic AI Agent Workflows

Varun Pratap Bhardwaj

Autonomous AI agents are deployed at unprecedented scale, yet no principled methodology exists for verifying that an agent has not regressed after changes to its prompts, tools, mo…

cs.CR2026

Formal Analysis and Supply Chain Security for Agentic AI Skills

Varun Pratap Bhardwaj

32 pages, 5 theorems with full proofs, 68 references, open-source tool: https://github.com/qualixar/skillfortify. v2: corrects the bibliography (22 entries had author lists that di…

cs.AI2026

Agent Behavioral Contracts: Formal Specification and Runtime Enforcement for Reliable Autonomous AI Agents

Varun Pratap Bhardwaj

Traditional software relies on contracts -- APIs, type systems, assertions -- to specify and enforce correct behavior. AI agents, by contrast, operate on prompts and natural langua…

cs.AI2026

SuperLocalMemory: Privacy-Preserving Multi-Agent Memory with Bayesian Trust Defense Against Memory Poisoning

Varun Pratap Bhardwaj

We present SuperLocalMemory, a local-first memory system for multi-agent AI that defends against OWASP ASI06 memory poisoning through architectural isolation and Bayesian trust sco…