thinkies-research-and-teach

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a purely instructional workflow (Scope, Research, Structure, Teach, Persist) focused on educational outcomes.
  • [EXTERNAL_DOWNLOADS]: The README.md provides installation links to the author's own GitHub repository (synapseradio/ai-skills). These are legitimate vendor resources used for distribution and updates.
  • [INDIRECT_PROMPT_INJECTION]: The skill's research phase (Step 2) uses tools to fetch external web content. This is a common design for research agents but introduces a theoretical surface where malicious instructions on a third-party website could attempt to influence the agent's output.
  • Ingestion points: External websites retrieved via web fetching tools in Phase 2.
  • Boundary markers: None explicitly defined in the skill instructions to delimit fetched content.
  • Capability inventory: The agent is restricted to research and text generation/teaching within this skill scope.
  • Sanitization: No specific sanitization or filtering of fetched content is defined; the skill relies on the agent's internal safety guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 09:28 AM
Security Audit — agent-trust-hub — thinkies-research-and-teach