understand-domain
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands for path resolution and environment detection, including git commands and realpath. It also executes a local Python script, extract-domain-context.py, to analyze the project codebase.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and processes arbitrary source code and metadata files (like README.md) from the project directory. While the script performs truncation for token management, it lacks explicit sanitization or boundary markers when providing this context to a subagent for domain analysis.
Audit Metadata