Security audit
Security checks for vulnerabilities and agentic risk
This is a prompt-engineering guidance skill made of documentation, with no hidden execution, persistence, or credential handling found.
Before installing, treat this as a documentation-heavy prompt-engineering helper. Share only prompts, files, or documentation you are comfortable giving to the model, especially when using examples involving detailed reasoning, linked references, or file-based analysis.
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.
- Explain methodology before results
### Don'ts
- Don't make predictions beyond 12 months without caveats
- Don't ignore outliers without investigation
- Don't present correlation as causation
- Don't use jargon without explanation
This is a markdown file, so SQP-2 applies to omitted warnings in skill descriptions. The section recommends eliciting explicit step-by-step reasoning but provides no caution about limiting such traces when handling sensitive, private, or confidential inputs, which could affect data exposure in outputs.
This markdown file suggests replacing full prompt context with links to detailed docs, which can affect privacy or system integrity if those references contain sensitive or inaccessible material. The guidance does not include any warning to verify that linked content is appropriate to share, accessible to the model, or safe to expose.
No suspicious patterns detected.