teach
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to "open the lesson file for the user by running a CLI command." This grants the agent authority to execute shell commands on the host machine, which could be exploited if file paths or parameters are manipulated by untrusted data.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it generates interactive content using data from potentially untrusted files in the workspace.
- Ingestion points: Untrusted data enters the context from
MISSION.md,RESOURCES.md, andlearning-records/*.mdwhen the agent reads them to prepare lessons. - Boundary markers: Absent. The format specifications do not include delimiters or instructions for the agent to ignore embedded commands within the processed data files.
- Capability inventory: The skill performs file writes to create HTML, CSS, and JS components in
./lessons/and./assets/, and it is authorized to execute CLI commands to open these files (SKILL.md). - Sanitization: Absent. There are no instructions to sanitize or validate content fetched from external resources or stored in workspace files before it is included in generated lessons.
Audit Metadata