Show HN: MCP server for AI compliance documentation (Colorado AI Act)(github.com)
github.com
Show HN: MCP server for AI compliance documentation (Colorado AI Act)
https://github.com/jeremytuite/aop-mcp-server
1 comments
Happy to answer questions from anyone testing it.
The core loop is simple: every request through the gateway gets a trace_id, a trust score, and a signed decision context. The audit trail stays in your own infrastructure — no external
calls.
One thing we're actively working on: calibrating trust thresholds per risk category under EU AI Act Article 13. If anyone's done similar work on confidence scoring for high-risk AI
systems, I'd genuinely like to compare notes.
Repo + 5-minute quickstart: https://github.com/base76-research-lab/cognos-proof-engine
The problem: organizations deploying AI in hiring, lending, insurance, or healthcare decisions need specific documentation — risk management policies, impact assessments, consumer notifications, bias testing docs, and appeal mechanisms. Most teams either pay $50K+ for a GRC platform, hire a law firm at $500/hr, or wing it.
What I built: compliance protocols that are both human-readable (PDF) and agent-readable (structured JSON via MCP/CLI/API). Your AI assistant can check if you're a deployer, pull protocol schemas, and help you implement them.
Tools available via MCP: - colorado_ai_act_check — are you a deployer? - list_protocols — browse by vertical - get_protocol_schema — structured format for agent implementation - assess_compliance — gap analysis
Install: npx -y aop-mcp-server
The Colorado AI Act is the first state-level AI governance law with teeth ($20K/violation, AG enforcement). More states are coming.
Site: https://appliedoperationsprotocols.com