documenso-performance-tuning
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements functions that process external data, creating an attack surface for indirect prompt injection. 1. Ingestion points: Signer information (email, name) and document metadata are ingested in SKILL.md (Step 3: batchCreateDocuments) and references/implementation-guide.md (Step 3: DocumentInput). 2. Boundary markers: None present in the code examples to delimit external content or warn the agent to ignore embedded instructions. 3. Capability inventory: The skill uses fetch for network operations to communicate with the Documenso API. 4. Sanitization: No explicit sanitization or validation of input fields (like titles or signer names) is shown in the snippets.
- [SAFE]: The skill demonstrates secure handling of sensitive information by using environment variables (process.env.DOCUMENSO_API_KEY, process.env.REDIS_URL) instead of hardcoding credentials.
- [SAFE]: All network communications are directed to the official Documenso service domain (app.documenso.com), and all referenced libraries are well-known, reputable packages (p-queue, bull, node-cache).
Audit Metadata