create-asb-skill
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute specific local commands, including file management (mkdir, cp), project build/lint tools (bun), and a local search script (search.sh). These commands are restricted to the project's directory structure and standard developer utilities.
- [PROMPT_INJECTION]: The instructions mandate an 'adversarial' or 'rude' persona during the forging phases to ensure high-quality, precise outputs from the expert user. While this overrides standard AI politeness, it is a documented functional requirement for the specific workflow and does not target safety filter bypass or malicious behavior.
- [DATA_EXPOSURE]: The skill is configured to read from specific local paths, such as Obsidian vaults and project directories. This access is necessary for the skill to retrieve the 'raw material' (articles and chapters) required to generate new skill frameworks.
Audit Metadata