press-release-writing
Pass
Audited by Gen Agent Trust Hub on Mar 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the
infshCLI tool (inference.sh) via Bash to perform research tasks, login to services, and run specialized AI applications for fact-checking. - [EXTERNAL_DOWNLOADS]: The documentation instructs the user to install additional skill sets from
inference-sh/skillsusing thenpx skills addcommand, which fetches external code at runtime. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8).
- Ingestion points: Search results retrieved from Tavily and Exa via the
infsh app runcommands (SKILL.md). - Boundary markers: The skill does not define specific delimiters or instructions to ignore malicious prompts embedded within the retrieved search data.
- Capability inventory: Access to Bash and the
infshCLI tool for executing commands and managing authentication (SKILL.md). - Sanitization: No explicit sanitization or validation of the external search content is mentioned before it is processed by the agent.
Audit Metadata