Grounded Cognitive Architecture for Enterprise AI Agents: Eliminating Hallucination via Mandatory Tool Execution at Scale
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The source code for the Virtual-Enterprise Platform is proprietary and not publicly available. However, we provide extensive supplementary materials to support reproducibility and verification of all claims. Production metrics (Prometheus data, Grafana dashboards, anonymized agent logs) are available to researchers upon request with appropriate confidentiality agreements. For academic collaborations and research validation, interested researchers should contact the corresponding author (amit.patole@gmail.com). https://doi.org/10.5281/zenodo.19253980 Supplementary materials include: Generalizable Framework Specification — complete technology-agnostic architecture (cognitive loop, ethics enforcement, multi-agent collaboration) suitable for independent implementation (4000+ lines) Production Metrics Dataset — agent quality grades, tool call success rates (98.4%), LLM routing distribution, pulse frequency data across 80 deployed agents Claim Verification Report — systematic verification of every empirical claim in this paper against production database state Evaluation Scripts — A/B comparison scripts (cognitive vs. legacy mode), agent population verification, reproducible against any PostgreSQL deployment Database Schemas — SQL migrations implementing the cognitive architecture, escalation rules, and temporal attack pattern detection tables Architecture Diagrams — 10 system diagrams covering all layers from cognitive loop to infrastructure Ethics Screening Evidence — temporal pattern detection examples (gradual escalation, fragmentation, jailbreak injection) with hallucination monitoring methodology and findings Implementation Guide — 12-month adoption roadmap for researchers and practitioners seeking to replicate the architecture
Fields of Study & Subjects
Artificial Intelligence Applications
Psychiatry
Mental Health
Neuroscience
AI-based Problem Solving and Planning
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 27 March 2026
- Language
- EN