airblackbox avatar

Air Blackbox

airblackbox/air-blackbox-mcp

Gives Claude 14 tools to scan Python AI agents for EU AI Act compliance across six articles, with the option to remediate violations instead of just reporting them. The basic install runs a lightweight scanner that checks code strings, files, or entire projects, explains regulatory requirements, detects prompt injection patterns, and generates compliance reports. Install the full SDK and you get GDPR scanning, bias analysis, pre-execution action validation gates, and compliance history tracking. Works with LangChain, CrewAI, AutoGen, and other agent frameworks. Under the hood it uses air-trust's HMAC-SHA256 audit chain with Ed25519 signatures. You can optionally wire it to a local Ollama model for deeper analysis beyond regex patterns. Reach for this if you're building AI agents in Europe or need to document regulatory compliance before deployment.

GitHub stars
1
tools
14
transport
STDIO
last updated
Feb 27, 2026

Tools

14 tools

Public tool metadata for what this MCP can expose to an agent.

scan_codeScan a code string for all 6 EU AI Act articles (Python, notebooks, TypeScript, JavaScript, Java, Kotlin with the SDK; Python without).

Scan a code string for all 6 EU AI Act articles (Python, notebooks, TypeScript, JavaScript, Java, Kotlin with the SDK; Python without).

No parameter schema in public metadata yet.

scan_fileRead and scan a single source file.

Read and scan a single source file.

No parameter schema in public metadata yet.

scan_projectRecursively scan a project directory.

Recursively scan a project directory.

No parameter schema in public metadata yet.

analyze_with_modelDeep analysis via local fine-tuned model (Ollama).

Deep analysis via local fine-tuned model (Ollama).

No parameter schema in public metadata yet.

check_injectionDetect prompt injection attacks (15 patterns).

Detect prompt injection attacks (15 patterns).

No parameter schema in public metadata yet.

classify_riskClassify tools by EU AI Act risk level.

Classify tools by EU AI Act risk level.

No parameter schema in public metadata yet.

add_trust_layerGenerate trust layer integration code.

Generate trust layer integration code.

No parameter schema in public metadata yet.

suggest_fixGet article-specific fix recommendations.

Get article-specific fix recommendations.

No parameter schema in public metadata yet.

explain_articleTechnical explanation of EU AI Act articles.

Technical explanation of EU AI Act articles.

No parameter schema in public metadata yet.

generate_compliance_reportFull markdown compliance report.

Full markdown compliance report.

No parameter schema in public metadata yet.

scan_gdprGDPR-specific compliance scan.

GDPR-specific compliance scan.

No parameter schema in public metadata yet.

scan_biasBias and fairness analysis.

Bias and fairness analysis.

No parameter schema in public metadata yet.

validate_actionValidate agent actions before execution (Article 14).

Validate agent actions before execution (Article 14).

No parameter schema in public metadata yet.

compliance_historyView past scans, trends, and compliance scores.

View past scans, trends, and compliance scores.

No parameter schema in public metadata yet.

Sponsors

Readme

AIR Blackbox MCP Server

AIR Blackbox MCP demo

EU AI Act compliance scanning for Claude Desktop, Claude Code, Cursor, and any MCP-compatible client.

Unlike other compliance scanners that only report problems, AIR Blackbox also remediates - generating working code fixes, trust layer integrations, GDPR compliance checks, bias analysis, and full compliance reports. Under the hood, the scanning feeds into air-trust, a cryptographic audit chain (HMAC-SHA256) with Ed25519 signed handoffs that ensures compliance data integrity.

14 Tools (10 base + 4 SDK-powered)


View the full README on GitHub

Details

Repository
github.com/airblackbox/air-blackbox-mcp
Package
air-blackbox-mcp
Transport
STDIO
Registry
MCP registry, active