guard
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local bash scripts located in the user's home directory (e.g.,
~/.claude/skills/gstack/...) to implement safety hooks and manage configuration state. - [INDIRECT_PROMPT_INJECTION]: The skill captures a directory path from the user and interpolates it into a shell command.
- Ingestion points: User input requested via the
AskUserQuestiontool inSKILL.md. - Boundary markers: No explicit instructions to ignore instructions within the path are provided to the agent.
- Capability inventory: Execution of bash commands and scripts via the
Bashtool. - Sanitization: The user-provided path is enclosed in double quotes in the resulting shell command.
- [SAFE]: The skill records local telemetry by writing the current git repository name and timestamp to a local file in the user's home directory. No network exfiltration or sensitive data access was identified.
Audit Metadata