prompt-engineering
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional markdown files and templates for prompt engineering. Comprehensive analysis found no evidence of prompt injection, obfuscation, or malicious intent designed to bypass safety filters or compromise systems.
- [SAFE]: The technical templates included in the skill (e.g., in
templates/technical-prompt-template.md) explicitly advocate for security best practices. For example, the 'Code Review' template includes mandatory checks for SQL injection, XSS vulnerabilities, and secret exposure. - [SAFE]: All external references and dependencies mentioned in the documentation, such as Cloudflare, PostgreSQL, and Grey Haven (the author's own resource), are to well-known, trusted technology services or official vendor resources.
- [SAFE]: The skill promotes the use of boundary markers and delimiters (Principle 8) as a core structural guideline. This teaching serves as a primary defense against indirect prompt injection by ensuring the model clearly distinguishes between instructions and user-provided data.
- [SAFE]: The skill is configured with a restricted set of file system tools (Read, Write, Grep, Glob, TodoWrite) and contains no network-capable tools or instructions for external data exfiltration.
Audit Metadata