AgentShield v1.1.0

AgentShield v1.1.0 — Deterministic AI Agent Security & Governance Framework

AgentShield v1.1.0 introduces our official security portal, interactive security studio, OWASP 2026 benchmark proof matrix, declarative defense catalog, and full documentation suite.

🌐 Official Website & Live Tools

• Website Home: https://codesphere-u.github.io/agentshield/
• Interactive Security Studio & MCP Generator: https://codesphere-u.github.io/agentshield/studio.html
• OWASP 2026 Benchmark Proof Matrix: https://codesphere-u.github.io/agentshield/proof.html
• Defense Rules Matrix: https://codesphere-u.github.io/agentshield/rules.html
• Full Architecture & Integration Docs: https://codesphere-u.github.io/agentshield/docs.html

What's New in v1.1.0

• Deterministic AI Agent Firewall: Zero-external-API runtime inspection across prompt injection, tool abuse, untrusted context poisoning, and sensitive data leakage.
• OWASP 2026 Alignment: Comprehensive coverage against OWASP Top 10 for LLM & Agentic Applications with 100% detection recall across 25 benchmark scenarios.
• Interactive Studio: Live policy testing, payload simulator, and custom Model Context Protocol (MCP) server configuration generator.
• Production Hardened: Sub-millisecond latency (<0.15ms per evaluation), zero third-party telemetry, SHA-256 hash-chained tamper-evident audit ledger.

AgentShield v1.0.0 — OWASP 2026 AI Agent Security & Governance Framework

Initial flagship release: multi-layer prompt injection detector, capability policy firewall (Shell, Filesystem, SSRF, SQLi), egress secret redactor, SHA-256 hash-chained audit ledger, 25-scenario OWASP benchmark suite, CLI & FastAPI Security Lab.