aprender
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input during study sessions, including exercise answers and code completions. This data is used to adapt the educational flow and is recorded in local state files.
- Ingestion points: Student responses provided during activities defined in ACTIVITIES.md and SKILL.md.
- Boundary markers: The skill instructions in GUIDED-CODE.md use code comments to delimit student implementation sections.
- Capability inventory: Filesystem read/write access to the local state directory (~/.kipperacademy/aprender/) and vendor-specific MCP tool calls for profile management and event recording.
- Sanitization: The skill relies on checking student answers against pre-defined criteria in PLAN.md rather than explicit string sanitization.
Audit Metadata