og-image-design
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
infshcommand-line tool to run various image generation and search applications on the inference.sh platform. These commands are consistent with the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: The skill references the installation of additional tools from the
inference-sh/skillsrepository using thenpx skills addcommand. These are vendor-controlled resources intended for extending the agent's capabilities. - [PROMPT_INJECTION]: The skill's use of the
html-to-imageapplication involves rendering raw HTML provided as input. This creates a surface for indirect prompt injection if the agent interpolates untrusted data into these HTML templates. - Ingestion points: Input JSON strings in
SKILL.mdcode blocks. - Boundary markers: None present in the examples.
- Capability inventory: CLI execution of
infshfor image rendering and search. - Sanitization: No sanitization or escaping of input data is described in the templates.
Audit Metadata