skills/scottwater/skills/solo/Gen Agent Trust Hub

solo

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill functions as an orchestration layer that spawns sub-agents and processes based on incoming prompts. This design represents an indirect prompt injection surface where malicious instructions could be passed through to spawned agents.
  • Ingestion points: Prompts and task lists passed to the solo_task tool as described in SKILL.md.
  • Boundary markers: The skill includes specific instructions to the agent to monitor sub-agents for permission requests and to only approve those necessary for the prompt, serving as a procedural boundary.
  • Capability inventory: The skill utilizes the Solo MCP toolset (solo_task, solo_scratchpad, solo_todo) to spawn processes and manage project-scoped state.
  • Sanitization: The skill describes the use of an acceptance: "none" flag to bypass user-in-the-loop gates when the calling skill is assumed to provide its own validation, which reduces oversight of the sub-agent's actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 12:15 AM
Security Audit — agent-trust-hub — solo