learn
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill uses local shell commands (
mkdir,while [ -f ... ],command -v ao) to manage directory structures and check for local CLI tools. These are standard operations for file-based knowledge management. - [PROMPT_INJECTION] (LOW): The skill processes free-text input from the user to generate file names and content. While it includes logic to sanitize slugs (removing special characters), the content itself is stored without sanitization.
- [INDIRECT PROMPT INJECTION] (LOW): This skill presents an attack surface where malicious instructions could be saved into the knowledge base.
- Ingestion points: User input via
/learn [content]orAskUserQuestion(SKILL.md, Step 1). - Boundary markers: None. The content is written directly into a markdown file with minimal frontmatter (SKILL.md, Step 5).
- Capability inventory: Uses
mkdir -p,while -f, andcommand -vvia shell (SKILL.md, Step 3, 4, 6). - Sanitization: The skill specifies removing special characters for the slug but does not specify sanitization or escaping for the content before writing to disk.
Audit Metadata