upskilling-coach
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
greputility to search local documentation files in thereferences/directory. This is used to retrieve specific coaching strategies based on the context of the user's learning struggles. - [PROMPT_INJECTION]: The skill processes untrusted user data regarding learning goals and current progress levels. This data is used to determine which coaching patterns to apply and what topics to search for in the reference files.
- Ingestion points: User responses to diagnostic questions in Phase 1 and follow-up check-ins.
- Boundary markers: None identified. User input is interpolated into the workflow logic.
- Capability inventory: Uses shell command execution (
grep) to query local files. - Sanitization: There are no instructions for sanitizing or escaping user-provided topics before they are used as patterns in shell commands, creating a surface for potential command injection if the agent passes raw user input to the search tool.
Audit Metadata