Latest

  1. Prompt injection examples and how to stop indirect attacks

    How prompt injection and indirect prompt injection steer AI agents, four documented examples, and the layered controls that limit the damage.

    Daniel Begimher

    Testing
    9 min
  2. Five days, five security incidents that broke the internet.

    PHP2Shell is broadly exploited, LiteLLM 1.107.0 shipped malware (49,000 downloads), and OpenCode, Railway and eMudhra put keys at risk. What to do now.

    Daniel Begimher

    Notes
    5 min
  3. How to get an AI agent approved: 12 questions reviewers ask

    What security, risk and compliance teams ask before they approve an AI agent in a regulated company, and the evidence that answers each question.

    Daniel Begimher

    Standards
    9 min
  4. Mapping AI agent risks to MITRE ATLAS and the NIST AI RMF

    A crosswalk from the OWASP agentic Top 10 to MITRE ATLAS techniques and mitigations and NIST AI RMF outcomes, and how to write findings reviewers accept.

    Cristian Leo

    Standards
    5 min
  5. AI penetration testing for agents in regulated industries

    What a penetration test of an AI agent should cover at a bank, insurer or hospital, how to run it safely, and the evidence that gets the agent approved.

    Daniel Begimher

    Testing
    6 min
  6. AIUC-1 and ISO 42001 for AI agents: what auditors ask for

    ISO/IEC 42001 checks how a company governs AI. AIUC-1 checks a specific agent. What each standard asks for, and the evidence your agents need for both.

    Daniel Begimher

    Standards
    5 min
  7. Armorer Guard: 3.4ms Inline Prompt Injection Defense

    Armorer Guard runs inline at agent boundaries where scanner latency becomes response latency. Latest benchmark: 3.4ms average local scans, 977 cases.

    Cristian Leo

    Research
    7 min
  8. Securing OpenClaw with Armorer Guard

    OpenClaw needs a runtime boundary. Armorer wraps it with managed setup, Docker hardening, health checks, approvals, audit trails, and Guard-backed scanning.

    Cristian Leo

    Runtime
    2 min

About this blog

Notes on secure agent runtimes, open-source AI infrastructure, and Armorer product work.

Armorer Labs writes here about running AI agents safely in production. Posts cover how to test and approve agents in regulated industries, prompt injection defense, and how we benchmark detection speed and accuracy. Each guide links to its primary sources, such as regulators, standards bodies and MITRE, so reviewers can check every claim. For incidents and new research on attacks against AI agents, see Threat Intel.

Stop collecting findings. Start closing them.