prompt-engineering
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation resource for prompt engineering best practices and contains no executable code or dangerous instructions.
- [SAFE]: Code snippets provided in the 'Degrees of Freedom' and 'Integration Patterns' sections are illustrative examples of how to structure prompts and templates, rather than commands for the agent to execute as part of the skill's own functionality.
- [SAFE]: The 'Persuasion Principles' section describes psychological techniques such as Authority, Commitment, and Scarcity used to improve instruction adherence in LLMs. This is presented as educational guidance for building reliable agent workflows and does not contain malicious injection attempts.
Audit Metadata