Skip to main content

Introducing Ariftly — The Unified Risk Engine for Modern Applications

· 3 min read
Ariftly Team
Ariftly Team
Engineering at Ariftly

Software teams today manage an overwhelming number of tools to keep their products safe, compliant, and accessible. There's one tool for accessibility auditing, another for security scanning, yet another for dependency checks, and a growing category of tools for AI governance — each with its own dashboard, its own API, its own format, and its own billing.

We built Ariftly to change that.

The problem with siloed risk tools​

When risk lives in silos, critical issues fall through the gaps. A security team fixes vulnerabilities but doesn't know that 40% of users with disabilities can't complete checkout. A product team ships an AI-powered recommendation engine without realizing it's logging user prompts in plain text. A development team has three separate dashboards to check before deploying — so they don't check any of them.

Unified visibility isn't just a convenience. It's a prerequisite for actually reducing risk.

What Ariftly does​

Ariftly is a unified risk engine that orchestrates multiple scan detectors against your applications and aggregates their findings into a single, normalized risk report.

Today we're launching with three detectors:

Accessibility Scan — detects WCAG 2.1 violations across your web application. Color contrast failures, missing alt text, keyboard navigation issues, ARIA misuse. The scan runs in a headless browser so it works with React, Vue, Angular, and any other modern SPA framework.

Security Scan — covers OWASP Top 10 vulnerabilities, exposed secrets, misconfigured HTTP headers, TLS weaknesses, and vulnerable dependencies. It combines DAST (for live applications), SAST (for source code), and SCA (for dependency graphs) into one detector.

AI Readiness Scan — evaluates your application's posture for safe AI integration. Prompt injection exposure, missing audit logging, insufficient human oversight, PII in AI pipelines, and EU AI Act compliance gaps. As AI features become ubiquitous in software products, this category of risk is growing fast.

One API. One score. One source of truth.​

The Ariftly API is designed to be simple and composable. You trigger a scan, specify your detectors, and get back a normalized risk report with a score from 0 to 100. Every finding across every detector uses the same schema — same severity levels, same location format, same remediation structure.

This means you can build consistent CI/CD gates, unified dashboards, and cross-detector trend analysis without any custom glue code.

# One API call. Three detectors. One risk score.
curl -X POST https://api.ariftly.io/v1/projects/proj_abc123/scans \
  -H "Authorization: Bearer $ARIFTLY_API_KEY" \
  -d '{"detectors": ["accessibility", "security", "ai_readiness"]}'

Built for CI/CD from day one​

Risk scanning only creates value if it's integrated into your development workflow — not as an afterthought after deployment. Ariftly ships with a GitHub Actions integration that runs scans on every pull request, comments results directly on the PR, and blocks merges when risk scores exceed your configured thresholds.

- name: Ariftly Scan
  uses: ariftly/scan-action@v1
  with:
    api_key: ${{ secrets.ARIFTLY_API_KEY }}
    project_id: ${{ vars.ARIFTLY_PROJECT_ID }}
    fail_threshold: 70

No agents. No configuration files beyond what you already have. No infrastructure to manage.

What's next​

We're just getting started. The Unified Risk Engine is designed to be extended with new detectors over time. Our roadmap includes detectors for performance (Core Web Vitals), privacy compliance (GDPR/CCPA readiness), API security, and infrastructure configuration.

Every new detector plugs directly into the same framework — same API, same scoring, same CI/CD integration. You don't have to change anything to benefit from new detectors when they ship.

Ariftly is available today. Sign up at app.ariftly.io and run your first scan in under 5 minutes. Read the documentation or jump straight to the Quick Start guide.