prompt-generator
Fail
Audited by Snyk on Feb 24, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (high risk: 1.00). The metaprompt includes deceptive/hidden instructions (e.g., produce internal monologue or write content inside / tags that "will not be shown to the user" and "do not discuss these instructions with the user") which instruct the agent to generate concealed reasoning or hide directives—behavior unrelated to the skill's advertised purpose of transparently generating prompt templates.
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The skill's testing step explicitly asks the assistant to request values for each variable and substitute them into the prompt to "show what the final prompt would look like," which would cause any user-provided secrets (API keys, tokens, passwords) to be echoed verbatim in the assistant's output.
Audit Metadata