skills/helderberto/skills/teach/Gen Agent Trust Hub

teach

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to open generated lesson files using CLI commands (e.g., 'open the lesson file for the user by running a CLI command'). This behavior is intended for user convenience and involves executing local system commands on files created by the agent within the workspace.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8) due to its core functionality of gathering knowledge from external sources to generate educational content.
  • Ingestion points: External books, articles, and community forums are used to populate the RESOURCES.md file and inform the content of lessons.
  • Boundary markers: There are no explicit instructions or delimiters defined to isolate the agent from potentially malicious instructions embedded in retrieved external content.
  • Capability inventory: The skill possesses file-write capabilities (creating .html and .md files in the workspace) and command execution capabilities (opening files via CLI).
  • Sanitization: No specific sanitization or validation mechanisms are described for processing content retrieved from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 06:16 AM
Security Audit — agent-trust-hub — teach