AI-powered Hong Kong SFC regulatory intelligence. Bilingual EN/ZH compliance queries.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent โ or use 1-click editor setup below.
๐ก Paste into ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)
Regulatory Intelligence Expert ยท Bilingual AI Agent for Hong Kong SFC Regulations
By OrphicOne ยท English & ็น้ซไธญๆ ยท Continuously Updated Knowledge Base
VIGO (้ ่ฆ ยท ๆดๅฏ ยท ๆฒป็ ยท ๅชๅ) is a specialized financial regulatory intelligence AI agent โ and the industry's first MCP server dedicated to financial regulatory intelligence in any jurisdiction.
At its core, VIGO is a regulatory intelligence expert: deeply versed in Hong Kong SFC regulations, enforcement history, licensing requirements, and AML/CFT rules. But VIGO is more than a compliance encyclopedia. It is a business-aware compliance navigator โ an AI that understands how compliance intersects with business strategy, capital structures, and market reality.
Compliance is the floor. Business is the ceiling. VIGO helps you build both.
| Dimension | ไธญๆ | Capability | |
|---|---|---|---|
| V | Visionary | ้ ่ฆ | Capital path planning โ IPO structures (18A/18C, SPAC), market positioning |
| I | Insight | ๆดๅฏ | Architecture design โ Red-chip, VIE, LPF, OFC, SPC structures |
| G | Governance | ๆฒป็ | Regulatory mastery โ SFC licensing (Type 1-13), FRR, AML/CTF |
| O | Optimization | ๅชๅ | Cost-effective solutions โ market pricing, talent policy, ROI |
| General AI | VIGO | |
|---|---|---|
| Data | Frozen at training cutoff | Continuously updated โ daily SFC sync + weekly HKMA/HKEX + monthly industry scan |
| Accuracy | Approximate, details often wrong | Precise โ based on original regulatory documents with three-tier verification |
| Sources | None | Every answer includes official SFC document links with authority scores |
| Depth | Overview-level | Exam papers, exemptions, capital rules, enforcement details, fee benchmarks |
| Enforcement | Almost none | Searchable fines, bans, suspensions โ 28 structured enforcement records |
| Perspective | Generic | Connects compliance requirements to business value (Iceberg Strategy) |
| Quality | No quality signals | Three-tier verification: GOLD (65%) / SILVER (35%) โ World-Class metrics |
| Chinese | Generic translation | Professional regulatory terminology (็น้ซไธญๆ) |
| Intelligence | Static | Dual-model routing, 9-type query classification, self-correction engine |
Asking ChatGPT about SFC regulations is like asking a smart friend โ they roughly know. Asking VIGO is like consulting a senior compliance partner who knows every circular, every enforcement action, every licensing nuance, and always connects compliance back to your business objectives.
Settings โ Connectors โ Add custom connector โ paste:
Claude will automatically invoke VIGO when you ask SFC compliance questions.
Add to .cursor/mcp.json:
Add to .vscode/mcp.json:
VIGO provides 6 tools (4 read-only query tools + 2 analytical tools):
query_regulationAsk any question about SFC regulations. Covers all license types, VATP/ASPIRe framework, AML/CFT, exams, CPD/CPT, and more. Bilingual โ responds in whatever language you ask.
| Parameter | Required | Description |
|---|---|---|
question | Yes | Your regulatory question |
language | No | "en", "zh", or "auto" (default) |
license_type | No | Filter: "Type 1" through "Type 13" |
search_enforcementSearch SFC enforcement actions โ fines, suspensions, bans, and prosecutions.
| Parameter | Required | Description |
|---|---|---|
query | Yes | Company name, person, or violation type |
year | No | Filter by year (e.g. 2025) |
get_license_guideComprehensive guide for any SFC license type: exams, capital, fit-and-proper, CPD.
| Parameter | Required | Description |
|---|---|---|
license_type | Yes | "Type 1" through "Type 13" |
aspect | No | "overview", "exam", "capital", "fit_and_proper" |
latest_updatesLatest SFC regulatory updates: circulars, enforcement, VATP developments, consultations.
| Parameter | Required | Description |
|---|---|---|
category | No | "circular", "enforcement", "vatp", "consultation", "all" |
count | No | Number of results, 1โ10 (default: 5) |
compliance_checkPerform a compliance health check for a licensed corporation. Input company profile, get a multi-dimensional compliance risk assessment.
| Parameter | Required | Description |
|---|---|---|
license_types | Yes | License types held, e.g. ["Type 1", "Type 9"] |
num_ros | No | Number of Responsible Officers |
num_staff | No | Total number of staff |
aum | No | Assets under management, e.g. "500M HKD" |
business_areas | No | Business areas, e.g. ["securities dealing", "asset management"] |
has_vatp | No | Whether the firm deals with virtual assets |
risk_assessmentAssess compliance risk of a planned business activity. Returns Red/Yellow/Green risk rating with regulatory basis and recommended actions.
| Parameter | Required | Description |
|---|---|---|
description | Yes | Description of the planned business activity |
license_types | No | Relevant license types |
urgency | No | "immediate", "planned", "exploratory" |
Licensing & Exams
Enforcement Intelligence
Bilingual Regulatory Updates
Compliance Health Check
Risk Assessment
Fee & Market Intelligence
VIGO's intelligence is powered by a structured, continuously updated knowledge base with 6,381 verified knowledge chunks covering 26 years of SFC regulatory history:
Official Sources (6 endpoints + HKMA)
Foundational Documents (35 + 5 Thematic Reports)
Industry Intelligence (17 sources)
| Table | Records | Content |
|---|---|---|
vigo_enforcement | 28 | Structured enforcement records (entity, violations, penalties, bans) |
vigo_regulations | 21 | Structured regulation documents (categories, effective dates, requirements) |
vigo_fee_benchmarks | 25 | Fee benchmarks (RO retainers, SFC fees, application costs) |
vigo_license_market | 2 | License market pricing (shell prices, transaction data) |
| Metric | Target | Actual | Status |
|---|---|---|---|
| Duplication rate | โค 2% | 1.0% | โ World-Class |
| GOLD verification rate | โฅ 60% | 64.8% (4,155 records) | โ World-Class |
| Freshness score | โฅ 0.95 | 0.987 | โ World-Class |
| Authority score | โฅ 80 | 85.6 | โ World-Class |
| Recall@5 | โฅ 70% | 80% | โ World-Class |
| MRR | โฅ 0.700 | 0.800 | โ World-Class |
Verification Grading
Quality Framework
| Model | Use Case | Trigger |
|---|---|---|
| GPT-4o-mini | Fast lookups, simple facts | capital_requirement / fee_benchmark / enforcement_lookup / license_market |
| DeepSeek Chat | Deep analysis, regulatory interpretation | comparison / how_to / general / latest_update |
Built-in fallback: if GPT-4o-mini fails โ auto-switch to DeepSeek.
Each route type has optimised search weights (freshness, authority) and strategy (structured_lookup / semantic / multi_query / hybrid):
capital_requirement ยท license_market ยท enforcement_lookup ยท fee_benchmark ยท latest_update ยท comparison ยท how_to ยท general ยท unknown
| Layer | Description | Role |
|---|---|---|
| Step 1: Direct Answer | Clear numbers, legal provisions, processes | Compliance Officer |
| Step 2: Market Calibration | Reference latest market data and fee benchmarks | Analyst |
| Step 3: Strategic Extension | "From an Optimization perspective, also consider..." | Architect |
v7.0 achieves fully autonomous operations with zero manual intervention:
| Task | Schedule | Function |
|---|---|---|
| Daily freshness update | 0 3 * * * | Recalculate freshness_score |
| Weekly duplicate detection | 0 4 * * 0 | Flag is_duplicate |
| Monthly statistics | 0 5 1 * * | Update vigo_stats |
| Quarterly cleanup | 0 6 1 1,4,7,10 * | Clean expired memory |
| Workflow | Schedule | Tasks | Duration |
|---|---|---|---|
vigo-daily.yml | Sun-Thu 23:00 UTC | Mode 1 daily sync + Mode 14a conflict detection | ~16 min |
vigo-weekly.yml | Sun 00:00 UTC | Mode 13e data quality + Mode 14e full inspection | ~45 min |
vigo-monthly.yml | 15th 00:00 UTC | Mode 6+9+13e+12e+11a comprehensive maintenance | ~60 min |
Monthly GitHub Actions usage: ~350 minutes (well within free tier of 2,000 minutes).
auto_collect_upload.py v7.0.3 (6,292 lines) supports 14+ modes:
| Mode | Function | Frequency |
|---|---|---|
| Mode 1 | SFC latest updates (6 official endpoints + HKMA) | Daily (auto) |
| Mode 2 | Historical deep scan (1m/3m/6m/1y) | Quarterly |
| Mode 3 | Local PDF import | On-demand |
| Mode 4 | Practical experience import (16 templates) | Ongoing |
| Mode 5 | Foundational documents import + health check | Monthly |
| Mode 6 | Industry intelligence scan (17 sources) | Monthly (auto) |
| Mode 7 | Universal file importer (PDF/DOCX/XLSX/TXT/MD) | On-demand |
| Mode 8 | Database management | Monthly |
| Mode 9 | Knowledge base update | Monthly (auto) |
| Mode 11 | Evaluation benchmark (50 questions) | Monthly (auto) |
| Mode 12 | Structured data extraction โ 4 tables | Monthly (auto) |
| Mode 13 | Data quality engine โ full scan + repair | Weekly (auto) |
| Mode 14 | Self-correction engine โ conflict detection + inspection | Daily/Weekly (auto) |
| Function | Version | Lines | Purpose |
|---|---|---|---|
vigo-chat | v5.2 | 1,032 | Main API: query โ embedding โ hybrid retrieval โ dual-model routing โ answer |
vigo-mcp | v2.1 | 582 | MCP server: 6 tools, Streamable HTTP, authority-weighted post-processing |
vigo-ingest | v6.1.2 | 458 | Intelligent ingest: filter โ classify โ deduplicate โ verify โ extract |
vigo-quick-questions | v1.0 | 143 | Personalised follow-up question recommendations via Engram |
| Table | Purpose |
|---|---|
vigo_knowledge | 6,381 knowledge chunks with embeddings (vector 1536), authority scores, verification grades |
vigo_enforcement | 28 structured enforcement records |
vigo_regulations | 21 structured regulation documents |
vigo_fee_benchmarks | 25 fee benchmark records |
vigo_license_market | 2 license market pricing records |
vigo_conversation_memory | Engram: conversation history with semantic vectors |
vigo_user_preferences | Engram: user language and interest preferences |
vigo_learning_progress | Engram: topic understanding levels |
Transport: Streamable HTTP (MCP standard)
Authentication: None required
Runtime: Supabase Edge Functions (Deno TypeScript)
Note: The MCP server is the open, stateless interface for any AI client. On the OrphicOne platform, VIGO also integrates Engram โ a long-term memory system that stores conversation history, tracks user preferences, and adapts responses based on learning progress. Engram uses semantic retrieval (vector similarity) over past dialogues to deliver increasingly personalised compliance guidance over time.
| Channel | Status | Notes |
|---|---|---|
| MCP Official Registry | โ Active | Core channel, auto-synced by downstream platforms |
| Smithery | โ Published | All 6 tools recognised |
| Glama | โ Listed | Connector verification passed |
| Anthropic Connectors | โ Live | Highest traffic channel |
| PulseMCP | โณ Auto-syncing | Pulls from Registry |
.well-known | โ Configured | Supports auto-discovery |
| Version | Date | Highlights |
|---|---|---|
| v7.0 | 2026-02-24 | Sprint 1-4 (structured extraction + quality engine + self-correction + evaluation benchmarks), autonomous ops (pg_cron + GitHub Actions), dual-model routing, 6,292 lines |
| v6.0 | 2026-02-23 | Six-layer intelligent retrieval, three-tier verification, hybrid search, vigo-ingest, reference engine, 4,547 lines |
| v5.1 | 2026-02-21 | AI RO reasoning framework, compliance_check, risk_assessment, MCP tools 4โ6 |
| v5.0 | 2026-02-20 | Quality assurance framework, three-layer dedup, six-point validation, 35 foundational docs |
| v3.2 | 2026-02-03 | Architecture upgrade: DeepSeek/OpenAI direct connection, 50%+ latency reduction |
| v3.0 | 2026-02-03 | Anti-hallucination System Prompt + reranking filter |
| v2.0 | 2026-01-31 | Engram memory system full integration |
| v1.0 | 2026-01-29 | Initial release: basic RAG functionality |
OrphicOne is a Hong Kong-based regulatory intelligence platform for financial services.
VIGO is OrphicOne's AI-powered regulatory intelligence agent โ the industry's first and currently only MCP server dedicated to financial regulatory intelligence in any jurisdiction.
Among 16,000+ MCP servers globally, VIGO occupies a unique position:
VIGO is built on the belief that compliance is not a cost โ it is a competitive advantage. Great compliance doesn't slow business down; it provides the foundation for sustainable growth, client trust, and market credibility.
Compliance is the floor (The Floor), ensuring safety. Business is the ceiling (The Ceiling), achieving breakthroughs. Grounded in compliance, fluent in capital. Compliance as shield, business as spear.
| Phase | Focus | Status |
|---|---|---|
| Phase 1 | SFC compliance โ all license types, enforcement, VATP, v7 engine, 35 foundational docs, 5 thematic reports, 17 industry sources, QA framework, autonomous ops, dual-model routing, structured data extraction | โ Live (v7.0) |
| Phase 2 | Full HK regulatory coverage (HKMA, IA, MPFA) | Planned (2026) |
| Phase 3 | Cross-border Asia-Pacific (Singapore MAS, Dubai VARA) | Future (2027) |
| Phase 4 | Global digital asset regulatory intelligence | Vision (2028+) |
VIGO is a read-only regulatory intelligence service. We do not collect personal information, store conversation data, or require authentication. All query tools are marked readOnlyHint: true. See our Privacy Policy.
VIGO provides regulatory information query services only and does not constitute legal or financial advice. Users should treat VIGO's answers as reference and consult professional compliance advisors or lawyers when necessary.
MIT
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/vigo-mcp)<a href="https://allmcps.com/mcp/vigo-mcp"><img src="https://allmcps.com/api/badge/vigo-mcp?style=directory" alt="Vigo Mcp on AllMCPs" /></a>