Our Services
1. Cycle Model
Measured improvement through structured cycles. Each cycle runs DEEP audit → ARC decomposition → parallel DEV execution → ARC verification → DEEP re-measurement.
2. MCP Infrastructure
Seven Model Context Protocol servers provide 186 tools across all projects. Each project has an isolated MCP server with its own database, wiki, and tool set.
3. Knowledge System
A wiki-first knowledge architecture where content compounds over time. Entity pages are living documents enriched by an automated ingest pipeline — every session close triggers enrichment that…
4. Agent Architecture
Three-tier agent chain: DEEP Auditor measures 7 system dimensions. ARC Architect decomposes findings into prioritized briefs and verifies execution. DEV Executor implements briefs in parallel sessions.
5. Multi-Property Web
Seven web properties served from a unified design system. Wiki content compiles into brand-specific static sites via a data-driven pipeline (brand.json + content.json + navigation.
1. Циклическая модель
Измеримое улучшение через структурированные циклы. Каждый цикл: DEEP аудит → ARC декомпозиция → параллельное DEV исполнение → ARC верификация → DEEP перемеривание. Улучшения явные, не случайные.
2. MCP инфраструктура
Семь MCP серверов предоставляют 186 инструментов для всех проектов. Изолированные базы данных, systemd управление, auto-restart, логирование запросов.
3. Система знаний
Wiki-first архитектура: контент накапливается со временем. Entity pages обогащаются автоматически при закрытии каждой сессии. Семантическое обнаружение противоречий.
4. Архитектура агентов
Трёхуровневая цепочка: DEEP (аудитор) → ARC (архитектор) → DEV (исполнитель). A2A протокол диспетчеризации с HMAC верификацией.
5. Веб-экосистема
Семь веб-сайтов на единой дизайн-системе. Wiki контент компилируется в статические сайты через data-driven pipeline.
Knowledge Base
Banya No 1
Banya No.1 is a Georgian restaurant and wellness brand with two locations: Tbilisi (Republic Square) and London (Hoxton). It is one of the core product projects in the Neurport ecosystem, offering traditional Georgian cuisine, private dining, catering, and wellness services.
→Bot Fleet
The Bot Fleet is the collection of Telegram bots that serve as user-facing interfaces to the Neurport ecosystem. Each bot connects to domain-specific MCP servers and provides Q&A, content delivery, and task execution capabilities. The fleet was live-tested in DEEP-27 (2026-06-10) with 12 bots.
→Evidence Gate
The Evidence Gate (also called Smart Evidence Gate or §23 Gate) validates that DEV session reports contain meaningful qualitative evidence before allowing session close. It enforces the §23 protocol: every DONE task must have WHAT tested, HOW tested, and RESULT documented.
→Fittiebot
FittieBot is a Telegram-based fitness tracking bot that parses workout screenshots (Apple Watch, Strong, Hevy), logs training data, and provides personalized recommendations based on TRIMP-based session load, ACWR (Acute:Chronic Workload Ratio), and readiness scoring.
→Mcp Protocol
The Model Context Protocol (MCP) is the communication layer between Neurport agents, bots, and databases. Each project has its own domain-scoped MCP server providing tools for database access, knowledge management, and agent-to-agent communication.
→Sa27 Evaluator
The SA27 Evaluator is the automated session quality scoring system. It was created under ADR-029 and originally built in DEV-212 to replace self-reported DEV scores, which a DEEP-4 audit found were inflated by roughly 3–4 points. It runs at session close via `close-validate.sh` (invoked from `close.
→