documenso-cost-tuning
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements logic to track and optimize Documenso usage patterns. It performs network requests to
app.documenso.com, which is a well-known service relevant to the skill's stated purpose. Authentication is securely managed using theDOCUMENSO_API_KEYenvironment variable. - [INDIRECT_PROMPT_INJECTION]: The skill includes functionality to process document metadata (such as document status, titles, and recipient details) fetched from the Documenso API to generate usage statistics. This represents a potential surface for indirect prompt injection, though the risk is negligible as it is restricted to metadata processing for the skill's primary optimization purpose.
- Ingestion points: Document data is retrieved using the
client.documents.findV0method in bothSKILL.mdandreferences/implementation-guide.md. - Boundary markers: The provided code snippets do not include explicit boundary markers or instructions to ignore instructions embedded within metadata.
- Capability inventory: The skill's capabilities are limited to making authorized network requests to the Documenso API and outputting aggregated statistics via
console.log. - Sanitization: The implementation examples do not demonstrate explicit sanitization of document metadata before it is processed or logged.
Audit Metadata