Research, release, and technical evidence archive
2026.03 Snapshot baseline for architecture, evaluation, and integration teams that need release notes, protocol papers, benchmark context, and technical references in one archive.
public-safe test evidence
core reasoning and orchestration
from trust edge to control plane
services and support systems
Primary releases and papers to read first
This is not a publication feed. It is an archive ordered for teams that need to understand platform evolution, protocol layers, and technical evidence.
v2.5.0: Infrastructure Layer - Containerization, API Gateway & Monitoring
Production-grade containerized infrastructure with API Gateway, monitoring, load testing, OWASP hardening, and chaos engineering scenarios.
v2.4.0: Analysearch Intent Engine & 41-Algorithm Framework
A release focused on the capability tiers behind RCT reasoning, orchestration, verification, memory, and autonomous planning.
v2.3.1: Enterprise Hardening - Security, Validation & Resilience
Layer 10 hardening across validation, injection protection, caching, circuit breakers, bulkhead isolation, and retry strategies.
v2.3.0: Control Plane - JITNA Wire Schema, Cryptographic Signing & Replay
Layer 9 covering the JITNA packet protocol, cryptographic signed execution, and deterministic replay with integrity verification.
SignedAI: Multi-LLM Attestation with 0.3% Hallucination Rate
Consensus-based AI verification designed for regulated environments, with a measured 0.3% hallucination rate and full auditability.
RCTDB v2.0: Persistent Enterprise Memory Architecture
A persistent memory architecture for enterprise agents, combining structured context, provenance, compression, and cross-session continuity.
JITNA Protocol RFC-001 v2.0: The HTTP of Agentic AI
Formal specification for AI-to-AI communication, intent specification, negotiation lifecycle, wire schema, and universal adapters.
Where to continue after each paper
Structured follow-on routes keep the research archive distinct from the blog and the evaluation hub.
Read for buyer evaluation
Start with comparative analysis and benchmark framing before procurement or governance reviews.
Read for protocols and systems
Continue into the protocol layer and architecture references if your team is planning integration or control-plane work.
Read for trust and evidence context
Use the benchmark summary and methodology pages to connect papers to the public-safe trust narrative.
Use this archive as a routing layer, not a dead end
After reading a release or paper, continue into evaluation, methodology, benchmark summary, and integration pages to complete the enterprise decision path.