avoid-ai-writing

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The README and PROOF reproduction instructions instruct users/agents to git-clone or curl the skill from https://github.com/conorbronsdon/avoid-ai-writing and raw GitHub URLs (e.g. https://raw.githubusercontent.com/conorbronsdon/avoid-ai-writing/main/SKILL.md and the cursor-rules raw URL), which causes remote text (the SKILL.md / rule files) to be fetched and injected as agent instructions at runtime—an external content fetch that directly controls agent behavior.