detect-ai-text
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill interacts with an external service at
https://www.stipple.shto perform text analysis. This is the primary function of the skill and is clearly documented. - [DATA_EXFILTRATION]: Document content is uploaded to a third-party API for analysis. The skill documentation includes a dedicated 'Limitations and Safety' section that transparently informs users about sending data to a hosted third party and advises removing confidential material before use.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from various document formats (PDF, DOCX, TXT, MD) and URLs. It uses structured transmission (multipart form data or JSON via
curl) to send this data to the detector, minimizing the risk of local command injection or prompt confusion during the ingestion process.
Audit Metadata