kanban-ai

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • COMMAND_EXECUTION (LOW): The skill relies on several bundled shell scripts (view_board.sh, search_by_tag.sh, etc.) to process markdown files. While these scripts use common utilities like grep, awk, and sed, they operate on local files in a user-defined directory. The use of bash <SCRIPTS_DIR>/... is a standard execution pattern for this type of skill.\n- PROMPT_INJECTION (LOW): As a tool that reads and displays content from user-generated markdown files, it is theoretically susceptible to indirect prompt injection if a card contains instructions meant to manipulate the agent's behavior when viewing the board or searching content. However, the scripts primarily perform string extraction and formatting, which limits the impact.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:41 PM