capsule
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill uses the
$ARGUMENTSvariable to ingest user-provided descriptions of a project or feature. This creates a surface for indirect prompt injection where a malicious input could attempt to redirect the agent's behavior during the documentation process. However, the skill provides a strict output schema and a structured interview process which mitigates this risk. - Ingestion points: User-provided project area defined in
$ARGUMENTSwithinSKILL.md. - Boundary markers: Absent; input is placed directly after the header 'Input'.
- Capability inventory: The skill instructs the agent to create or update files specifically within the
docs/directory (e.g.,docs/capsule.md). - Sanitization: None; the skill relies on the structured output format and validation steps to ensure the agent stays on task.
Audit Metadata