learn
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of shell commands (including
ghq,find,ln,mkdir,sed, andrm) to manage local repository structures, create symlinks, and automate cloning workflows. - [EXTERNAL_DOWNLOADS]: Fetches external repository content from GitHub using the
ghqcommand. This is a core functionality of the skill for exploring remote codebases. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted source code from external repositories using parallel AI agents.
- Ingestion points: External code is read from the
origin/symlink directory as defined inSKILL.md. - Boundary markers: Absent; the instructions to the agents do not include delimiters or specific warnings to ignore instructions embedded within the source code files.
- Capability inventory: The skill possesses capabilities for filesystem management (
mkdir,ln,rm,unlink) and repository cloning (ghq). - Sanitization: No sanitization or filtering is performed on the content of the external source code before it is passed to the AI agents for analysis.
Audit Metadata