learn-teach
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input via the
$ARGUMENTSplaceholder.\n - Ingestion points: User-provided concepts are ingested at the end of the
SKILL.mdfile.\n - Boundary markers: The input is not wrapped in delimiters or accompanied by instructions to ignore embedded commands.\n
- Capability inventory: The skill contains no executable scripts, allowed-tools, or subprocess calls, limiting the risk to purely conversational manipulation.\n
- Sanitization: No validation or sanitization of the
$ARGUMENTScontent is performed.
Audit Metadata