teach
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to fetch and process data from external URLs (identified in
RESOURCES.md) to generate lessons. This architecture is vulnerable to indirect prompt injection, where an attacker-controlled resource could contain instructions that influence the agent's behavior during content generation. - Ingestion points: External resource URLs and the
RESOURCES.mdfile. - Capability inventory: Writing files to the local filesystem (Markdown, HTML) and executing CLI commands to open files.
- Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the interpolation of external data.
- Sanitization: The instructions do not specify any validation or sanitization for content retrieved from external sources.
- [COMMAND_EXECUTION]: The skill directs the agent to execute a CLI command to open generated HTML lesson files (e.g.,
0001-slug.html) for the user to view. While intended as a user experience convenience, this involves the execution of shell commands based on generated file paths.
Audit Metadata