Building security infrastructure for AI agents.

Creator of SpiderShield and SpiderRating.

Security toolkit for AI agents and MCP servers.

• Static Scanner — 46 security issue codes, vulnerability detection, SARIF output, GitHub Action
• Runtime Guard SDK — tool-call firewall with policy enforcement (strict / balanced / permissive)
• MCP Proxy — transparent security proxy for MCP servers
• DLP Engine — PII and secret redaction for agent data flows
• Audit Trail — structured JSONL logging and local SQLite dataset
• Agent Security Checker — config audit, skill scanning, toxic flow analysis, allowlist mode

· GitHub

Security index and certification platform for AI agents and tool ecosystems.

• Rating Engine — certification scoring (F/D/C/B/A) with calibrated ground truth
• Enterprise Dashboard — multi-agent fleet security visibility and management
• Certification API — badge issuance, compliance reporting, EU AI Act readiness
• Advanced Analytics — behavioral baselines, anomaly detection, fleet intelligence

• AI Agent Security
• AI Infrastructure
• Financial AI Systems
• Prediction Systems

AI agents will interact with tools, APIs, and systems across the internet.

SpiderShield aims to provide the security infrastructure that makes this safe.

Los Angeles

spidershield.dev · PyPI