ado-create-story

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill implements defensive design by explicitly instructing the agent to ignore any arguments provided alongside the command, forcing the use of interactive prompts to collect data. This effectively mitigates common injection vectors via command parameters.
  • [COMMAND_EXECUTION]: The skill exclusively uses well-defined MCP tools (wit_add_child_work_items and wit_update_work_item) for its write operations. It does not use bash or other shell environments to execute arbitrary commands.
  • [DATA_EXFILTRATION]: While the skill reads project configuration from a local CLAUDE.md file, this data is used solely to target the correct Azure DevOps organization and project. No evidence of unauthorized data transmission or hardcoded credentials was found.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 12:48 PM