Search

Firewall for AI coding agents — intercept, log, approve every action.

Agent framework for remediating GitHub security alerts.

Security Trust Report: [email protected] — 54/100 (C, standard). Security incident detected. Maintainer risk, supply chain analysis from 8 security databases.

A visual, source-grounded guide from agent architecture to secure agentic AI systems.

Enterprise Security Control Plane SDK for Autonomous and Multi-Agent AI Systems

Always-on dependency security monitoring for DeepSeek Harness (DSH) plugins: find exact installed or candidate transitive vulnerable paths and breaking updates, then route evidence to a DSH Agent.

A local-first security testing platform for AI agents with adversarial scenarios, tool-call interception, approval workflows, budget controls, evidence-based findings, and external agent adapters.

Security scanner for AI agent tool definitions

AI model: shiqiangchen6/mcp-security-taxonomy

GitHub security posture analysis for AI agents — 39 MCP tools, 45 checks across org, repos, Actions, secrets, supply chain, and access control

Forensic reconstruction for AI agent security incidents (local-first, standards-aligned, defensive DFIR).

A safety layer that lets developers use AI coding agents at work without triggering security alerts

Portable, standards-backed security policies for any AI coding agent. One command to install OWASP, CWE, NIST rules + security skills.

SecureVector SDK for CrewAI — brings the local threat monitor's three controls (tool-call permissions, secret/data-leak detection, threat detection) to every CrewAI tool call, with tamper-evident audit logging.

Runtime risk scoring for AI agent tool calls. Detect privilege escalation, baseline deviations, and data exfiltration inline — under 1ms, zero dependencies, no LLM.

Interactive LLM agentic evaluation TUI for local and cloud models

Official Python SDK for AgentGuard Intelligence Engine and Action Firewall.

Guard LangChain.js tools with EMILIA Protocol — require an offline-verifiable authorization receipt (EP-RECEIPT-v1) before an irreversible tool runs: missing -> refused, valid -> runs, replay -> refused, forged -> refused (RR-1). The approval becomes port

Privacy-leakage testing for AI agents: audit the whole execution trace, score it deterministically, gate it in CI.

Static analyzer and GitHub Action for detecting Agentic Workflow Injection risks in AI-powered GitHub Actions workflows.