scaffold
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill analyzes local codebase files to determine patterns for scaffolding, creating an indirect prompt injection surface.\n
- Ingestion points: Reads existing source files (*.ts, *.py) and AGENTS.md to extract conventions (Steps 2 and 3).\n
- Boundary markers: No delimiters or safety instructions are used to distinguish project code from agent instructions.\n
- Capability inventory: The skill can write new files to the filesystem and execute shell commands (find, npx tsc, mypy).\n
- Sanitization: No sanitization or validation of ingested code content is performed.\n- [COMMAND_EXECUTION]: Executes shell commands to discover the project structure and verify the generated boilerplate.\n
- Evidence: Uses the find command to locate reference examples and verify directory placement.\n
- Evidence: Executes npx tsc and mypy to run type checking on the generated files.
Audit Metadata