multi-agent-workflow

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill is susceptible to indirect prompt injection because the agent is instructed to read and rely on status files generated by other potentially less-trusted agents.\n
  • Ingestion points: Reads progress-{agent}.md files as part of Step 3 in SKILL.md.\n
  • Boundary markers: Absent. There are no instructions or delimiters provided to ensure the agent ignores instructions embedded within these files.\n
  • Capability inventory: The agent has the ability to spawn other agents with arbitrary task descriptions using the oh-my-ag CLI and local scripts like spawn-agent.sh.\n
  • Sanitization: No sanitization or validation of the content within the progress files is mentioned.\n- [Command Execution] (SAFE): The skill requires the use of CLI tools (oh-my-ag) and local scripts (spawn-agent.sh) for agent orchestration. These are standard operations for the described project workflow and do not involve privilege escalation or unauthorized access within the context provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:49 PM