Security research, AI agent vulnerabilities, and best practices from the HackMyAgent team.
OpenA2A Shield combines credential protection, configuration integrity, runtime detection, and compliance scoring.
AI coding assistants read your .env files, terminal history, and MCP server configs. Here is how to protect credentials.
Run opena2a review in any project directory and get a security posture score with credential scanning and actionable fix commands.
OASB brings the CIS Benchmark model to agentic AI -- 46 controls, 10 categories, 3 maturity levels.
We contributed 6 security fixes to OpenClaw (205K+ stars). 4 PRs merged directly, 2 adopted by maintainers. Covers credential redaction, code safety scanning, path traversal, and more.
AI agents are making decisions, calling APIs, and accessing sensitive data autonomously. But most have no real identity. Here's how to give every agent a cryptographic identity.
OAuth 2.0 and OIDC power human authentication. But AI agents aren't humans. Here's the identity gap and how AIM solves it.
OASB defines the first comprehensive security benchmark for AI agents -- 46 controls across 10 categories with 3 maturity levels. Check compliance with HackMyAgent.
PR #9806 merged into OpenClaw -- 1,721 lines adding built-in skill security scanning. 6 checks detect malicious patterns at install and update time.
HackMyAgent v0.4.0 ships the first automated detection for CVE-2026-25253 (CVSS 8.8), expanded ClawHavoc IOCs, and 11 new security checks.
HackMyAgent is a security toolkit for AI agents with 4 modes: Attack, Secure, Benchmark, and Scan. Here's how to use each one.
We scanned 97,013 internet-facing hosts for AI agent vulnerabilities. 14.4% had confirmed security issues. Here's what we found.
Traditional NHI platforms manage service accounts and API keys. But AI agents are a fundamentally different class of non-human identity.
The ClawHavoc campaign planted 341 malicious skills on ClawHub. We built a scanner to detect it.
How each OWASP Agentic risk maps to NHI governance capabilities.
ServiceNow disclosed the most severe AI-driven vulnerability to date -- exposing 85% of Fortune 500 companies.
AIM provides cryptographic identity, MCP attestation, trust scoring, and audit logging for AI agents.
CVE-2025-32711 (EchoLeak) affected Microsoft Copilot. Learn how to secure your AI agents with AIM.