busa-ideas
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill takes untrusted user input describing product ideas and consultant answers and incorporates them into professional business documents.
- Ingestion points: User data enters the agent context through the
ideasandquestionsbases as defined inSKILL.mdand managed inbusabase-provider.js. - Boundary markers: The instructions for the agent to draft or revise documents do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing user input.
- Capability inventory: The skill hands off generated PRDs to
$kelly-app-skill-creatorfor application generation and performs data writes to the Busabase storage platform. - Sanitization: While the web UI component uses
DOMPurifyfor sanitizing rendered content, there is no explicit sanitization or instruction-filtering applied to the data before it is interpolated into the prompts for document generation or handoff.
Audit Metadata