skills/davidondrej/skills/teach/Gen Agent Trust Hub

teach

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: Analysis of the skill instructions and format definitions revealed no malicious patterns, obfuscation, or unauthorized data access.
  • [COMMAND_EXECUTION]: The skill instructions suggest running a CLI command to open generated HTML lesson files for the user. This is a functional convenience for a teaching agent and does not involve arbitrary or dangerous command execution.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and summarize external resources to create lessons.
  • Ingestion points: External URLs and community resources referenced in RESOURCES.md or retrieved via web search.
  • Boundary markers: Absent; the skill does not define specific delimiters to separate untrusted external content from its own instructions.
  • Capability inventory: Filesystem writes to create ./lessons/*.html and ./reference/*.html, and shell command execution to open these files.
  • Sanitization: Absent; the skill lacks instructions to sanitize or validate content fetched from external sources before generating output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 06:41 PM
Security Audit — agent-trust-hub — teach