api-testing

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The "Companion check" instructs the agent to run shell commands (ls ... and npx skills add ...) to enumerate and install local skills — a side-effecting, out-of-scope instruction unrelated to API testing and therefore a hidden/deceptive prompt injection.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.40). The skill is mainly a testing guide and safe, but its "companion check" explicitly instructs running shell commands (ls to inspect skill directories and an npx skills add ... install command) which would read and modify the local filesystem and install packages — it does not ask for sudo, modify system config files, or create users, so the risk is limited but non‑zero.