ComplyOS
AI-native compliance auditing engine for enterprise LMS, enabling querying compliance status, running audits, and validating assignment rules through natural language with MCP clients like Claude or Cursor.
README
ComplyOS
L&D Compliance & Learning Operations MCP Server
An AI-native compliance auditing engine for enterprise learning management systems. Built by someone who spent 12 years in L&D ops and got tired of explaining to regulators why the CSV export didn't match the dashboard.
Why ComplyOS?
Enterprise compliance tracking is a disaster of CSV exports, stale dashboards, and "I thought they completed that" moments. ComplyOS treats compliance as a first-class engineering problem:
- Evidence-backed audits — Every report includes a SHA256-hashed evidence ledger
- Assignment rule validation — Test targeting rules before they hit 10,000 users
- AI-native interface — Query status via Claude Code, Cursor, or any MCP client
- Local-first — SQLite by default; no SaaS lock-in
Architecture
┌─────────────────┐ ┌──────────────────┐ ┌─────────────────┐
│ MCP Client │────▶│ ComplyOS MCP │────▶│ Compliance │
│ (Claude/Cursor) │ │ Server (FastMCP)│ │ Auditor │
└─────────────────┘ └──────────────────┘ └────────┬────────┘
│
┌──────────────────────────────┼──────────────┐
│ │ │
┌──────▼──────┐ ┌──────────▼─────┐ ┌────▼─────┐
│ Mock │ │ Workday │ │ SAP/CSOD│
│ Connector │ │ Connector │ │ (planned)│
└─────────────┘ └────────────────┘ └──────────┘
Quick Start
Installation
# Clone the repo
git clone https://github.com/simongonzalezdc/complyos.git
cd complyos
# Install with uv (recommended)
uv sync --all-extras --dev
# Or with pip
pip install -e ".[dev]"
CLI Usage
# Run a compliance audit
complyos audit
# Filter by department
complyos audit --department Engineering
# Generate a structured report
complyos report --department Engineering --json
# Check a single user's status
complyos status u1
# Sync LMS data to local SQLite
complyos sync
# Validate an assignment rule before deploying
complyos validate-rule rule.json
# Preview who would be affected by a rule
complyos preview-rule rule.json
# Check connector health
complyos health
MCP Server
# Start the MCP server
complyos mcp
Then configure your MCP client (Claude Code, Cursor, etc.) to point to the server.
Connectors
| Platform | Status | Auth |
|---|---|---|
| Workday Learning | ✅ Supported | Basic Auth (env vars) |
| Mock (seed data) | ✅ Built-in | None |
| SAP SuccessFactors | 🚧 Planned | OAuth 2.0 |
| Cornerstone OnDemand | 🚧 Planned | API Key |
Workday Configuration
Set environment variables:
export WORKDAY_BASE_URL="https://your-workday-instance.com"
export WORKDAY_USERNAME="your-user"
export WORKDAY_PASSWORD="your-pass"
Development
# Run tests
uv run pytest -q
# Run with coverage
uv run pytest --cov=complyos --cov-report=term-missing
# Lint
uv run ruff check complyos tests
# Type check
uv run mypy complyos --ignore-missing-imports
Domain Model
ComplianceGap(
user=User(id="u1", department="Engineering", ...),
missing_courses=[Course(code="SEC-101", mandatory=True)],
severity=ComplianceGapSeverity.HIGH, # critical | high | medium | low
days_overdue=14,
remediation_action=RemediationAction(...),
)
Every audit produces an EvidenceLedgerEntry with SHA256 hashes for regulator-ready audit trails.
Roadmap
- [x] Phase 1 — Core auditor, MCP server, CLI, Workday connector, tests
- [x] Phase 2 — SQLite persistence, assignment rules engine, sync command
- [ ] Phase 3 — Remediation workflows, Web UI, Slack/Teams notifications
License
Apache-2.0
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。