Plimsoll: an agent skill for testing prompt injection, leaks, and tool abuse
Summary
Plimsoll is an open-source agent skill designed for red-teaming LLM applications and agents, focusing on security testing for issues like prompt injection, leaks, and tool abuse.
Similar Articles
POISE: Position-Aware Undetectable Skill Injection on LLM Agents
POISE is a stealthy skill-poisoning attack that embeds malicious triggers within benign-looking instructions, achieving high attack success rates while evading detection by LLM scanners.
I got paranoid about OpenClaw skills injecting crap into my system prompt, so I built a quarantine pipeline with two LLMs as reviewers (93.75% detection, zero false negatives)
A developer built a quarantine pipeline using two LLM reviewers (Claude and Codex) to detect injection attacks in OpenClaw skills, achieving 93.75% detection rate with zero false negatives. The system uses a dual mandate of checklist-based pattern matching and open analysis to catch both known and novel injection techniques.
Skill Inspector
Skill Inspector is a developer tool that audits AI agent skills to help prevent malware risks.
Agent Against Agent: An Agentic System for Automatic Prompt Injection Red Teaming
This paper introduces PIMiner, an agentic system for automatic prompt injection red-teaming that builds a strategy library during training and transfers to unseen target LLMs at test time, achieving strong attack success rates against models like Gemini-2.5-Pro, GPT-5.1, and Claude-Sonnet-4.5.
Security for your OpenClaw agent skill before they run
SecureSkill is a tool that performs 10-layer security analysis on OpenClaw agent skills before execution, detecting threats like credential harvesting, outbound calls, and shell scripts. It produces a signed audit report mapped to OWASP, MITRE, NIST, and EU AI Act standards.