Architect · Detect · Remediate
Pipeline modules
ariada.ai is the agentic compliance architect for the modern web. The platform decomposes into nine integrated pipeline modules — one Ariadne thread, woven across an Architect foundation tier, a Detect tier of six scanning modules, and a Remediate tier of three source-level remediation modules.
Each module is a focused capability that snaps into the platform. Marketplace standalones (blamer.ai, clamper.ai, reverter.ai) ship one module each; ariada.ai ships all nine under one contract.
01 Architect tier — foundation
L0 Mindset · L1 Design · L2 Dev Tools
The Architect tier is the foundation that transforms how client teams design, code, and ship — so the product is born with accessibility, security, and AI Act transparency as defaults, not patched in later. The Architect tier ships as an opensource framework (L0 Mindset), design-time plugins (L1 Design — Figma / Adobe / Sketch / Canva), and build-time tooling (L2 Dev Tools — ESLint plugins, MCP server, GitHub Action).
-
Module L0 Mindset
Opensource framework · 10 design rules · LMS
-
Module L1 Design
Figma / Adobe / Sketch / Canva plugins
-
Module L2 Dev Tools
IDE / ESLint / GitHub Action / MCP server
These are the Foundation tier — see Layer architecture for full L0–L10 detail.
02 Detect tier — six modules
Continuous measurement, multi-tool consensus, signed evidence.
Once a property is shipped, the Detect tier scans the deployed surface every cycle — one navigation, many domains, one canonical score, and a tamper-evident evidence chain a regulator can read.
-
Multi-domain scan
Single-pass multi-domain scan
-
Canonical scoring
Cross-tool canonical scoring
-
AI authorship attribution
AI authorship attribution
-
Regression detection
Regression detection
-
AI artifact audit
AI Artifact Audit (evidence ledger)
-
Scan visualization
Scan visualization
03 Remediate tier — three modules
Source-level pull requests. Human-in-loop. No runtime overlay.
Detection without remediation is just a complaint. The Remediate tier turns each finding into a source-level pull request — the client merges, ARIADA re-scans, the loop closes.
-
LLM remediation cascade
Tiered LLM remediation cascade
-
Backlog optimizer
MIP + ML backlog optimizer
-
CI/CD gate
CI/CD gate · clamper.ai
04 Partner ecosystem — not part of ARIADA
Regulator-facing dossier prep & submission.
ARIADA's nine modules cover the deployed-surface scan and the source-level remediation. They do not generate the regulatory dossier the DPA / notified body / FDA / MDR auditor reads — that is a separate workflow with a separate buyer. For that we partner with a sibling SaaS rather than build it ourselves.
-
Governancer
End-to-end EU AI Act · GDPR · CA SB 243 · NY GBL · Italian L. 132/2025 · MDR · FDA PCCP · NYC LL 144 dossier preparation and submission. Kernel + plugin SDK architecture; regulator-readable audit log export. Sibling SaaS under Agonist Development AB.
ARIADA scan reports flow into Governancer's audit log as evidence-chain entries (SDK GA Q3 2026; deep-link integration Q4 2026). See Partner ecosystem on /about for context.
Source-level remediation only — agentic suggestions are not autonomous deployments; pull requests require client merge. Not a legal certification body. For accredited certification, consult a notified body.