scratch
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary purpose is codebase exploration through hands-on execution in a dedicated .scratch/ directory. It incorporates explicit safety rules to prevent modification of production code, database writes, or configuration changes, ensuring a non-destructive analysis process.
- [EXTERNAL_DOWNLOADS]: The skill performs an update check by fetching a version file from the author's repository at https://raw.githubusercontent.com/abpai/skills/main/versions.json. This is a standard and transparent mechanism for checking for software updates from a known vendor.
- [COMMAND_EXECUTION]: The skill involves creating and executing local scripts (e.g., in TypeScript or Python) to exercise project modules. This execution is intended to help the agent and user understand code behavior through live execution and output capture.
- [PROMPT_INJECTION]: The skill does not contain any instructions that attempt to override system safety guidelines or redirect agent behavior toward malicious goals.
Audit Metadata