skill-vetter
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the shell_exec tool to run standard system utilities such as find, cat, and rm. It also invokes the skillhub CLI to install skills into a temporary directory (/tmp/skillhub-preview/) for inspection purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary operation involves reading and displaying untrusted text from external skills, creating a potential surface for indirect prompt injection if the content being vetted contains malicious instructions designed to hijack the agent.
- Ingestion points: Reads content from files in /tmp/skillhub-preview/ using the cat command (SKILL.md).
- Boundary markers: No specific delimiters or boundary warnings are implemented to isolate the untrusted content from the agent's context during the review process.
- Capability inventory: The agent has access to shell_exec to perform file listing, content reading, and directory removal (SKILL.md).
- Sanitization: The skill does not perform sanitization, escaping, or filtering of the ingested file content before it is processed.
Audit Metadata