skill-creator
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions explicitly direct the agent to execute and test newly generated or modified scripts to ensure they function as intended. This represents a controlled form of dynamic execution essential to the skill's purpose. Evidence: SKILL.md Step 4 ('Added scripts must be tested by actually running them').
- [COMMAND_EXECUTION]: The initialization script performs administrative filesystem operations, including setting file permissions to make scripts executable. Evidence: scripts/init_skill.py ('example_script.chmod(0o755)').
- [PROMPT_INJECTION]: The skill's workflow creates an attack surface for indirect prompt injection by ingesting user-defined requirements and script logic that are subsequently executed during the testing phase.
- Ingestion points: User-provided requirements and code implementation in Step 1 and Step 4.
- Boundary markers: None specified for the script content itself.
- Capability inventory: The agent is empowered to write to the filesystem (init_skill.py) and execute arbitrary scripts (test instructions).
- Sanitization: The validation script (quick_validate.py) checks metadata fields for length and restricted characters but does not sanitize or validate the executable script content.
Audit Metadata