content-repurposing
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation instructions for the inference-sh CLI tool hosted on the official GitHub repository for that project.
- [COMMAND_EXECUTION]: The skill utilizes the
beltCLI tool to perform various automated tasks such as generating images via AI models and posting content to social media platforms. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content (e.g., blog posts, podcast transcripts) for conversion, which creates a vector for indirect prompt injection attacks. 1. Ingestion points: External source content provided by the user for atomization (e.g., blog-to-thread conversion). 2. Boundary markers: The provided templates do not include specific delimiters or 'ignore' instructions to prevent the agent from following instructions embedded in the source content. 3. Capability inventory: The skill has access to the
beltCLI via Bash, which can interact with external APIs and social media platforms. 4. Sanitization: There is no evidence of input validation or content filtering mechanisms described in the skill.
Audit Metadata