writing-for-humans
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The instructions include a command to install a companion skill (
jd-solanki/ai) using thepnpm dlx skills@latestutility. This represents a standard dependency management operation within the platform for a tool provided by the same author. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and format user-generated content such as pull request bodies and commit messages. While this constitutes a data ingestion surface for untrusted text, the skill's instructions are focused on formatting and stylistic improvements, and it lacks dangerous capabilities (like network exfiltration or file system manipulation) that could be exploited via indirect injection.
Audit Metadata