Skills
skills/av/skills/timeboxed-iterating/Gen Agent Trust Hub

timeboxed-iterating

Warn

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to create a log file in /tmp with a name incorporating the user-provided 'Goal'. If the agent uses shell commands for this task, a malicious goal could execute arbitrary commands.- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface. Ingestion points: The 'Goal' input in SKILL.md. Boundary markers: Absent; the user-provided goal is interpolated directly into the subagent prompt without delimiters or 'ignore' warnings. Capability inventory: The orchestrator performs file operations and executes shell commands (date), while subagents are authorized to edit and commit repository code. Sanitization: Absent; the goal input is used without any filtering, validation, or escaping.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 4, 2026, 02:26 PM