skills/birdsigh/skills/suggest-next/Gen Agent Trust Hub

suggest-next

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and guides the AI's creative and analytical output. It does not contain any executable code, shell commands, or network operations beyond a recommendation to use the agent's built-in web search tool if available.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the conversation history and potentially from web search results to generate suggestions.
  • Ingestion points: Reads the full conversation history to detect context and surface ideas (SKILL.md).
  • Boundary markers: None explicitly defined in the prompt instructions to distinguish between conversation content and instructions.
  • Capability inventory: Purely text generation and information retrieval (web search).
  • Sanitization: None specified, though the output is restricted to a specific markdown structure which limits the impact of potential injections.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 03:12 PM
Security Audit — agent-trust-hub — suggest-next