classify_ai_systemLABSCO SUMMARY
One jurisdiction, followed all the way through.
Most AI-governance tooling is written against the EU AI Act and treats everything else as a footnote; this one is Canadian law end to end, from classification to the impact assessment. That makes it narrow on purpose — useful if AIDA is your regime, beside the point if it isn't.
What it is
A Python server for Canada's Artificial Intelligence and Data Act, covering impact assessment and algorithmic transparency. It turns the question "does AIDA apply to this system, and at what level" into something you can ask in a sentence.
What you get
Five tools: classify_ai_system sorts a system as high-impact or general-purpose under AIDA, impact_assessment runs the assessment AIDA Sections 7-8 call for, compliance_check tests it against AIDA's obligations, crosswalk_to_eu_ai_act maps those requirements to the EU AI Act, and generate_documentation drafts the compliance documents.
Requirements
Install with pip install canada_aida_ai_mcp; your client then starts it with the canada_aida_ai_mcp command, over stdio. MIT-licensed.
Setup effort
One command — pip install canada_aida_ai_mcp
USE CASES
Deciding whether AIDA covers a system you're shipping
Classify it first; the obligations follow from that answer.
Drafting the impact assessment
Generate the assessment from the system description rather than starting on a blank page.
Checking where you stand before a review
Run the compliance pass and see what is still open.
5 tools
Listed in project docs. Your client may show a different list.
impact_assessmentcompliance_checkcrosswalk_to_eu_ai_actgenerate_documentationBEFORE YOU INSTALL
What to know first
Tools you get
5 tools named in the project's documentation.
Requires
Python 3.10 or later
INSTALLATION
Connect your client
Add the block to your client, or run the command directly.
Server setup
Configuration file
{
"mcpServers": {
"canada-aida-ai-mcp": {
"command": "canada_aida_ai_mcp"
}
}
}pip
pip install canada_aida_ai_mcp