teach
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and explain untrusted external data including source code, git diffs, and debugger outputs. This creates a theoretical surface for indirect prompt injection where malicious instructions embedded within the processed data (such as code comments) could attempt to influence the agent's explanation or tone.
- Ingestion points: The skill instructions in
SKILL.mddirect the agent to 'Read the code yourself', 'Open the diff', and use the 'debugger'. - Boundary markers: The skill does not define specific delimiters or explicit 'ignore embedded instructions' warnings for the data it processes.
- Capability inventory: The skill can invoke sibling tools (
how,why,unslop) and use platform capabilities for Mermaid diagramming and image generation. - Sanitization: No explicit sanitization or filtering of external content is specified in the skill's instructions.
Audit Metadata