content-repurposing
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
infshCLI tool to perform AI-driven tasks such as image generation and text-to-speech. Commands likeinfsh loginandinfsh app runare standard for this tool's usage and are necessary for the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: Instructions guide the user to install the
infshCLI and add related skills from theinference-sh/skillsecosystem. These downloads are from the vendor's own infrastructure and are documented neutrally as they are expected dependencies. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes external content (e.g., blog posts or podcast transcripts) to generate social media output via tools like
x/post-create. 1. Ingestion points: The skill processes external content for conversion (SKILL.md). 2. Boundary markers: Not present in the provided templates. 3. Capability inventory: Includes social media posting and image generation. 4. Sanitization: No sanitization logic is shown in the examples. - [SAFE]: No obfuscation, hardcoded credentials, data exfiltration, or persistence mechanisms were detected in the skill content.
Audit Metadata