Status and limits

Works today

• Scanning a folder with 16 rules across the EU AI Act, GDPR, SOC 2 and ISO/IEC 27001
• The CLI, with Markdown and JSON reports and --fail-on for CI
• The MCP server with scan, classify, explain_framework and explain_rule
• Choosing frameworks with lintsure.config.yaml
• Outside-code lists with the evidence to collect
• A multi-page assessment report with a human review and sign-off page (--report)
• A GitHub Action that builds the engine from source and runs it, with a self-test
• A disclaimer and next steps on every report and every MCP reply
• Packages on npm: lintsure, @lintsure/mcp and @lintsure/core
• Public source code at github.com/lintsure/lintsure

Not yet

• Auditor-reviewed mappings. All four frameworks and every rule are in Beta until their mappings are reviewed.
• Deeper analysis. Rules match patterns in a line or a file. They don’t follow data across files or functions, so they can miss problems and sometimes flag safe code. Partial findings tell you when context matters.
• Languages. Rules target JavaScript, TypeScript and Python. Other languages are mostly not checked.
• A Fixer that writes code. Lintsure gives the fix as text; your coding agent makes the change.
• More frameworks. HIPAA, PCI DSS, ISO/IEC 42001 and others are on the roadmap, not in the engine.
• The team platform. Dashboards, history and evidence logs are coming soon.

Version

This documentation describes rules version 0.1.0.