teach
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and summarize external knowledge from URLs listed in
RESOURCES.md. An attacker could potentially influence the agent's behavior by placing malicious instructions in those external resources. - Ingestion points: The agent processes content from external links (RESOURCES.md) and local workspace files.
- Boundary markers: No specific delimiters or instructions to ignore embedded prompts in external data are provided.
- Capability inventory: The agent can write HTML/JS files and execute shell commands to open files.
- Sanitization: The skill lacks requirements for sanitizing content retrieved from external sources before it is incorporated into lessons.
- [COMMAND_EXECUTION]: The skill instructs the agent to 'open the lesson file for the user by running a CLI command', which involves direct shell execution to facilitate the viewing of generated materials.
Audit Metadata