Files

T

angel 72b6879f4b v1.7.0: refactor max_bot to flat structure, add VCF+UserModel+NLP history context, portal pages and proxy fixes

- Refactored max_bot from nested packages to flat module structure
- Q2: Extended BotUser model (patronymic, email, org, address, vcf_raw, contact_hash, phone_verified, email_verified, last_interaction, total_conversations, total_tickets)
- Q2: VCF parser (FN, N, TEL, EMAIL, ORG, ADR), upsert on re-contact, NLP history context (_get_user_history_context -> YandexGPT)
- Q1: Broadcast preview modal with 10s confirmation timer
- Q3: CSS var(--white)->var(--bg-card), var(--text)->var(--text-primary)
- Q4: bot_settings showNotification(), editable max_bot_id
- Q5: Webhook secret passthrough via X-Max-Bot-Api-Secret
- Masking sensitive keys, dialog_cleared handler, migrate via _add_column_if_not_exists()
- Rate limit (asyncio.sleep 0.5 per 10), dead code removed, conv.intent context in contact.py
- Portal pages: bot_consent, bot_kb (edit), bot_settings, bot_test, bot_tickets, portal_settings
- Tests: 21/21 passing, added test_yandex_gpt.py, test_email_sender.py
- Deploy: deploy_full.sh, schema.sql, seed_knowledge_base.sql

2026-05-29 02:30:30 +03:00

7.8 KiB

Raw Blame History

Ниже — единая цифровая операционная модель AegisOne Engineering в формате .md.

Это не просто документация — это “скелет системы”, который можно отдать разработчику, внедрить в Notion/Bitrix/1C/веб-CRM или развернуть как внутренний продукт.

AegisOne Engineering

Unified Digital Operating System (UDOS)

0. СУТЬ СИСТЕМЫ

AegisOne UDOS — это единая цифровая система управления инженерной эксплуатацией объектов безопасности.

Она объединяет:

CRM (клиенты и сделки)
SLA (контракты и обслуживание)
KPI инженеров (исполнение)
SHS (здоровье бизнеса)
Финансовую модель
Операционный дашборд

1. АРХИТЕКТУРА СИСТЕМЫ

┌──────────────────────┐ │ LEADS / CRM │ └─────────┬────────────┘ ↓ ┌──────────────────────┐ │ AUDIT SYSTEM │ └─────────┬────────────┘ ↓ ┌──────────────────────┐ │ SLA CONTRACTS │ └─────────┬────────────┘ ↓ ┌─────────────────────────────────────────┐ │ OPERATIONS (ENGINEERS) │ └─────────┬───────────────┬──────────────┘ ↓ ↓ ┌──────────────┐ ┌─────────────────────┐ │ KPI ENGINEERS │ │ INCIDENT SYSTEM │ └──────┬───────┘ └─────────┬───────────┘ ↓ ↓ ┌─────────────────────────────────────────┐ │ SHS ENGINE │ └──────────────┬──────────────────────────┘ ↓ ┌─────────────────────────────────────────┐ │ CEO DASHBOARD │ └─────────────────────────────────────────┘

2. CRM СИСТЕМА

2.1 Структура клиента

Client {
  id,
  company_name,
  object_type,
  location,
  contact_person,
  decision_maker,
  number_of_objects,
  systems: {
    video: int,
    access_control: int,
    fire_alarm: bool,
    it_infrastructure: bool
  },
  current_provider,
  pain_points[],
  budget_level,
  status: [lead, qualified, audit, offer, sla, lost]
}


---

2.2 Воронка CRM

Lead
 ↓
Qualification
 ↓
Technical Audit
 ↓
Risk Report
 ↓
SLA Offer
 ↓
Contract
 ↓
Active SLA


---

3. AUDIT SYSTEM (ИНЖЕНЕРНЫЙ ВХОД)

3.1 Формирование Risk Score

Risk Score =
  missing_archive (25)
+ no_power_backup (20)
+ no_regulations (15)
+ system_failures (20)
+ no_documentation (10)


---

3.2 Output аудита

Audit Report:
- Risk Score
- System condition
- Critical vulnerabilities
- SLA recommendation
- Cost estimation


---

4. SLA СИСТЕМА

4.1 Формула стоимости

SLA Price =
Base Cost × Object Index × Region Factor × Risk Multiplier


---

4.2 Object Index

Object Index =
(Risk × 0.4)
+ (Complexity × 0.3)
+ (Infrastructure × 0.2)
+ (Service History × 0.1)


---

4.3 SLA уровни

Class	Index	Description

A	0–30	simple SLA
B	31–60	standard SLA
C	61–90	complex SLA
D	90+	enterprise SLA



---

4.4 SLA метрики

Response Time

Resolution Time

Uptime %

Incident Rate



---

5. KPI СИСТЕМА ИНЖЕНЕРОВ

5.1 Engineer Score

ES =
(0.25 × SLA Compliance)
+ (0.20 × Response Time Score)
+ (0.20 × Resolution Time Score)
+ (0.15 × Diagnosis Accuracy)
+ (0.10 × Reopen Rate Score)
+ (0.10 × Documentation Quality)


---

5.2 KPI метрики

SLA Compliance

closed_in_SLA / total_requests

Reopen Rate

reopened_requests / total_requests

Diagnosis Accuracy

confirmed_faults / found_faults

Utilization

working_hours / available_hours


---

5.3 Грейды инженеров

Score	Level

90–100	Senior
80–89	Strong
70–79	Middle
<70	Junior



---

6. SHS (SYSTEM HEALTH SCORE)

6.1 Формула SHS

SHS =
(0.22 × SLA Stability)
+ (0.18 × Revenue Stability)
+ (0.18 × Retention)
+ (0.15 × Engineer Performance)
+ (0.12 × Incident Stability)
+ (0.10 × Sales Flow)
+ (0.05 × Operational Efficiency)


---

6.2 Интерпретация

SHS	Status

85–100	Growth
70–85	Stable
50–70	Risk
<50	Crisis



---

6.3 Alert System

SHS ↓ 5% weekly → Warning

SHS ↓ 10% → Critical

Retention drop → Immediate audit trigger



---

7. INCIDENT SYSTEM

7.1 Incident Classification

Level	Description

P1	critical system failure
P2	partial failure
P3	minor issue



---

7.2 Incident Rate

IR = incidents / number_of_objects


---

7.3 SLA breach tracking

breach_rate = failed_SLA_cases / total_cases


---

8. FINANCIAL LAYER

8.1 MRR

MRR = Σ SLA contracts / 12


---

8.2 Revenue per Engineer

RPE = SLA Revenue / number_of_engineers


---

8.3 Cost per Object

CPO = total_operational_cost / number_of_objects


---

9. CEO DASHBOARD

9.1 Main indicators

SHS (system health)

MRR (monthly recurring revenue)

SLA Compliance

Retention

Incident Rate

Engineer Score Avg

Sales Conversion



---

9.2 Dashboard visualization blocks

1. SHS Line Chart

trend over time


2. SLA Heatmap

engineers vs performance


3. MRR Growth Curve

monthly recurring revenue


4. Client Retention Funnel

drop-off points


5. Incident Map

problematic objects


6. Engineer Ranking Table

top/bottom performers



---

10. AUTOMATION RULES

10.1 If SHS drops

IF SHS < 70:
  reduce new sales
  trigger audit cycle
  reassign engineers


---

10.2 If SLA compliance drops

IF SLA < 90%:
  freeze non-critical projects
  increase inspection frequency


---

10.3 If retention drops

IF retention < 90%:
  mandatory client audit
  review engineer assignments


---

11. SYSTEM BEHAVIOR MODEL

DATA → KPI → SHS → DECISION → ACTION → FEEDBACK LOOP


---

12. STRATEGIC PRINCIPLE

AegisOne Engineering is not a service company.

It is:

> an operating system for managing engineering risk in physical infrastructure.




---

13. END STATE MODEL

The system evolves into:

predictive SLA risk detection

automated pricing

engineer auto-allocation

client churn prediction

MRR forecasting



---

END OF SYSTEM

---

Если хочешь следующий уровень, я могу собрать тебе:

1) визуальный UI дашборда (как он должен выглядеть как продукт)  
2) структуру базы данных (SQL / 1C / Bitrix schema)  
3) алгоритм “предикта потери клиента за 14 дней”  
4) или автоматический генератор КП + SLA из этой системы  

Это уже уровень **enterprise SaaS, а не просто инженерной компании**.

7.8 KiB Raw Blame History Unescape Escape