decision-mapping

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a template for stateful planning and investigation. It does not perform unauthorized network requests, credential harvesting, or privilege escalation.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to load a markdown file ("decision map") as context. While loading external files can theoretically introduce indirect instructions, the skill's design involves a highly structured workflow with human-in-the-loop handoffs, and it primarily processes data that it previously generated or that the user explicitly provides in the project directory.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes internal sub-skills such as /prototype, /grilling, and /domain-modeling. There are no patterns of downloading and executing untrusted scripts or packages from the internet.
  • [COMMAND_EXECUTION]: All operations involve standard file system interactions (reading and writing markdown files) and do not involve shell command injection or unsafe execution of user-supplied arguments.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 03:01 PM
Security Audit — agent-trust-hub — decision-mapping