skills/cursor/plugins/teach/Gen Agent Trust Hub

teach

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and process untrusted external data such as source code, diffs, and subsystem documentation to generate its explanations.
  • Ingestion points: The skill reads 'a body of work', 'code', 'diffs', and 'subsystems' from the user's project environment.
  • Boundary markers: The instructions lack specific boundary markers or delimiters to help the agent distinguish between its core instructions and the potentially untrusted content within the files it analyzes.
  • Capability inventory: The skill possesses the capability to read project files, access the debugger, and invoke other agent skills such as 'how', 'why', and 'unslop'.
  • Sanitization: There are no instructions for the agent to sanitize, filter, or validate the content it reads before incorporating it into its explanations.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 09:52 AM