nutrient-document-processing
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends using
npxto download and execute the@nutrient-sdk/dws-mcp-serverpackage from the public npm registry for its MCP server functionality. - [COMMAND_EXECUTION]: Provides several
curlcommand examples to interact with the Nutrient API, involving file system reads (@filename) and writes (-o output.pdf). It also usesnpxto run the vendor's MCP server. - [DATA_EXFILTRATION]: Transmits local document data (PDF, DOCX, etc.) to the external endpoint
api.nutrient.io. This is the documented primary function of the skill, and the destination is the official API of the service provider. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes data from untrusted external documents (e.g., OCR or text extraction).
- Ingestion points: Document files processed via
curlor the MCP server (referenced in SKILL.md). - Boundary markers: None identified; extracted text is not explicitly delimited in the provided instructions.
- Capability inventory: Subprocess execution via
curlandnpx, and file system access for reading/writing documents. - Sanitization: No sanitization or filtering of extracted text is mentioned before the agent processes the results.
Audit Metadata