5 papers
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,…
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…
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…
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…
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…