writing-for-agents
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is comprised solely of instructional Markdown documentation and basic YAML metadata.
- [NO_CODE]: No executable scripts or shell commands are present in the skill files. The analysis did not detect any Python or Node.js logic.
- [PROMPT_INJECTION]: The content provides advice on prompt engineering, such as using compact conceptual 'leading words' and avoiding negative steering. These instructions are intended to improve agent predictability and do not include patterns to bypass safety guardrails or hijack agent behavior.
- [DATA_EXFILTRATION]: There are no network requests, unauthorized file access patterns, or credential harvesting techniques present in the provided files.
Audit Metadata