data-ai MCP Server
Give your AI tools architecture-aware reasoning. Knowledge graph from any codebase — 14 LLM backends, impact analysis, confidence scores. Works with Claude Code, Cursor, VS Code Copilot.
Discovered via github-topic:mcp and last synced 3mo ago.
Install instructions not detected yet
Check the source repository for the latest setup steps.
Best For
Supply chain
Teach the graph — lesson persists across sessions and teams
Full 12-dimension UX friction audit with Claude Vision
Capture + optional Claude Vision analysis
Description
500-token focused context for any module
Touch targets + viewport
Auto-crawl all pages from a starting URL
3
**Graqle**
Confidence-gated prediction — writes back if threshold met
CSP, XSS, exposed API keys, auth flow
Click elements by text, selector, or coordinates
25,000
What GraQle ships
status`** — single UX entry-point for the EU AI Act mode toggle. `switch status` shows every EU-AI-Act-aware subsystem (Article 50 disclosure, Article 14 gate, claim-limits, baseline-doc, periodic-assessment, feedback-trend, EUR-Lex guard) in one envelope. - **Article 14 human-review enforcement** — `graq edit/apply/auto` refuses auto-apply when confidence < threshold (default 0.75, placeholder pending R25-EU-CALIB-01) AND EU AI Act mode is on. Structured refusal envelope with `article_14_clauses: ["14(4)(c)", "14(4)(d)"]`. - **R25-EU11 claim-limits v1.0** — typed vocabulary (17 canonical values, 6 categories) on every governance record. L08 SHACL + L19 audit-trail enforcement. Public attribution to Ricky Jones (TrinityOS). - **VERITAS Q16.1 baseline-doc generator** — `graq compliance baseline-doc generate` produces a dated, content-addressed artefact (SHA-256). Maps to EU AI Act Article 11 + ISO 42001 Cl. 6.2. - **VERITAS Q16.3 periodic-assessment** — `graq compliance periodic-assessment run` with auto-remediation triggers. Maps to Article 9 + ISO 42001 Cl. 9.1. - **VERITAS Q16.5 OBSERVATION-ONLY drift watcher** — `graq compliance feedback record/ingest` + Welford running statistics. Patent-novelty boundary enforced by mandatory AST audit test per Q-PATENT 2026-05-22. - **EUR-Lex weekly drift guard** — `graq compliance eur-lex-check` + GitHub Actions workflow re-fetches every cited EUR-Lex URL every Monday, opens issue on regulator-side drift. - **PCT (Proof-Claims Token) Use B** — `graqle.pct.issuer/validator` + `x-ai-eu` extension namespace (10 fields). First-public-draft of the OPSF `x-ai-eu` namespace authored by Quantamix. **Prior v0.56.0 surface preserved:** all 7 Article docs + CLI surfaces remain. Schema version stays at `"1"` — `graq compliance status --format json` is backward compatible (new `eu_ai_act_subsystems` field is additive). → [Full v0.57.0 changelog](./CHANGELOG.md#0570-2026-05-16---cr-010-eu-ai-act-wave-2) --- ## Pricing
No
What it does
Graph-of-agents reasoning — 50 nodes, multi-round, confidence-scored
Binary governance gate — PASS / FAIL with evidence
WCAG 2.1 accessibility
Type into forms, inputs, search boxes
Pre-publish gate
Blast radius — BFS traversal through dependency graph
Graph stats, node details, health status
Core Web Vitals
Run 10 audit dimensions on any live page
What it does
Various
Version bumps
Surface relevant past mistakes for current query
12 behavioral UX tests — zero AI cost
Open any URL, screenshot + full DOM summary
Session + auth profile management
Unlimited
Result
Rules-based
❌
Pre-change gate — lessons + safety boundaries + risk level
Autonomous loop — plan, generate, write, test, fix, retry until GREEN
CTA + trust signals
Execute multi-step user journeys with assertions