Skills
skills/ulpi-io/skills/run-parallel-agents-feature-build/Gen Agent Trust Hub

run-parallel-agents-feature-build

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes task lists and plans to generate briefs for sub-agents, creating a surface for indirect prompt injection. \n
  • Ingestion points: SKILL.md (Step 1: Analyze the Task List or Request). \n
  • Boundary markers: Absent; the skill relies on manual verification rather than programmatic delimiters. \n
  • Capability inventory: Agent tool for concurrent sub-agent execution, Skill tool for external tool invocation, and browse tool for browser sessions. \n
  • Sanitization: Absent; no explicit escaping or filtering of task content is described. \n- [COMMAND_EXECUTION]: Orchestrates external CLI tools codex and kiro-cli via the Skill tool for independent code reviews. This behavior is documented as a mechanism to obtain a 'second opinion' from separate AI systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 06:52 PM