og-image-design
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and fetches documentation from the inference-sh GitHub repository for CLI installation instructions.
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
beltCLI tool andnpxfor generating images and managing dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where HTML content is processed to generate images, creating a potential surface for indirect prompt injection if untrusted data is interpolated into the HTML templates.
- Ingestion points: HTML input strings within the
belt app run infsh/html-to-imagecommand parameters (SKILL.md). - Boundary markers: Absent; the examples show direct interpolation of text into HTML styles.
- Capability inventory: Execution of shell commands via the Bash tool as specified in the skill's allowed-tools (SKILL.md).
- Sanitization: No escaping or validation procedures are described for the HTML data being processed.
Audit Metadata