agent-code-explainer
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted code snippets through the
$ARGUMENTSvariable without implementing boundary markers or instructions to disregard potential commands embedded within that content. This creates an attack surface for indirect prompt injection. - Ingestion points:
SKILL.mduses the$ARGUMENTSplaceholder to receive external code. - Boundary markers: Absent; the skill does not utilize unique delimiters or specific instructions to the agent to treat the input as inert data.
- Capability inventory: The skill is assigned to the
general-purposeagent, which may have access to file system or network tools that could be exploited if the agent follows malicious instructions hidden in the input. - Sanitization: Absent; input data is directly placed into the instruction context without escaping or validation.
Audit Metadata