Search

AI Agent Security Competition SDK - Red teaming framework for tool-using AI agents

Security assessment framework for AI agents — adversarial test runner + server-side audit + scoring

Enterprise-grade Security Middleware for Multi-Agent AI Systems (LangGraph, LangChain) defending against Prompt Injections and Agent Exploitation.

Framework-agnostic agent security gateway — protect any AI agent with Veridex security packs, regardless of framework

A fail-closed gateway between an AI agent and its tools: normalize, default-deny policy, human approval, append-only audit.

Make consequential AI-agent actions safe to execute.

Intent-aware runtime governance and enforcement for autonomous AI agents interacting with APIs, tools, and MCP servers.

Autonomous Penetration Testing AI Agent

Universal cyber assistant: ~80 audited Kali tools + an expert skills library, driven by any model over MCP or direct run

Multi-tenant AI agent runtime with OS-level isolation. Sandboxed execution, encrypted credentials, private memory per tenant — one server, no Docker.

Nx plugin for [agent skills](https://agentskills.io/specification): any directory containing `SKILL.md` becomes a project with a full skill lifecycle — `build`, `lint`, `validate`, `os-check`, `size-check`. No `project.json` needed.

Check what an LLM is given for prompt injection: material at ingest, tool output, and the turn it answers

A reproducible, defensive red-team range for financial-services AI agents.

Contained static file analysis with Claude Code and Codex: REA, Ghidra and JADX in restricted Docker containers, selected input copies, readable reports. Linux/macOS; Windows experimental.

Official Vericto SDK for Python: check SQL before it runs. Wrappers for snowflake-connector-python, sqlite3, SQLAlchemy and any DB-API driver; integrations for LangChain, LlamaIndex, dbt and Airflow.

PolicyShield plugin for OpenClaw — runtime tool call policy enforcement

Map every integration point in your AI codebase — 13 CWE categories, attack surface score, test coverage gaps. SARIF 2.1.0. No API key.

AI model: hotdogs/Agents-A1-4B-cyber-lora (text-generation)

Alpha agent trust middleware: audit trails, safety guardrails, HITL, and tool validation

Adversarial assessment SDK for AI agents. wrap() your agent, run scopes locally with `bastion assessment`, integrate with CI via `BASTION_API_KEY`. Covers OWASP LLM Top 10.