grammarly-core-workflow-a
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to Grammarly's official API endpoint (api.grammarly.com). As a well-known service provider for writing assistance, this external communication is expected and safe within the context of the skill's functionality.
- [PROMPT_INJECTION]: The skill acts as a bridge for document text, creating an indirect prompt injection surface.
- Ingestion points: Untrusted text enters the system via the
scoreDocumentandbatchScorefunctions insrc/grammarly/scoring.ts. - Boundary markers: The provided code does not implement specific delimiters or instructions to the LLM to ignore instructions embedded within the processed document text.
- Capability inventory: The skill uses
fetchto send data to the external Grammarly API. - Sanitization: The implementation includes a check to ensure documents contain at least 30 words, adhering to API requirements.
Audit Metadata