skill-maker
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local scaffolding script (scripts/scaffold.py) to generate skill files. The generated skill templates also include instructions for the AI agent to execute a Python engine (scripts/run.py) for deterministic processing.
- [DYNAMIC_EXECUTION]: The skill generates new Python scripts and Markdown instruction files from hardcoded templates. This is standard behavior for a scaffolding tool and does not involve untrusted external data in the code generation process.
- [INDIRECT_PROMPT_INJECTION]: The tool ingests user-provided descriptions and triggers to populate generated skill templates. While it creates an attack surface by interpolating user input into new skill files, the script slugifies the primary name identifier to prevent directory traversal and the risk is contained within the generated skill context.
Audit Metadata