documenso-data-handling
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external document data and recipient information which could contain malicious instructions. While the skill advocates for security best practices like redaction, it does not explicitly define boundary markers to prevent the agent from obeying instructions embedded in the documents it processes.
- Ingestion points: Processes document metadata and content via the Documenso SDK (
client.documents.getV0,client.documents.findV0) in bothSKILL.mdandreferences/implementation-guide.md. - Boundary markers: The skill provides logic for sanitizing data for logs, but does not include explicit instructions for the agent to disregard content within the document data during runtime processing.
- Capability inventory: The skill utilizes capabilities for file system writes (
writeFile), network operations (fetchto Documenso API), and cloud storage operations (S3Client.sendto AWS S3). - Sanitization: Demonstrates PII redaction (email masking and name redaction) and provides encryption/hashing implementations (AES-256-GCM and HMAC-SHA256) to protect sensitive data.
Audit Metadata