data-analysis
Fail
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's knowledge base (references/knowledge-base.md) contains multiple links to domains that have been flagged by security scanners as malicious or associated with phishing. Specifically, reference [4] links to milagrosoft.com (Blacklisted) and reference [59] links to getmonetizely.com (Phishing). These resources pose a risk to the user and the agent if the URLs are visited or if content is fetched from them.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted data (e.g., CSV or database uploads) which could contain malicious instructions designed to manipulate the agent's behavior. 1. Ingestion points: Dataset uploads processed in the 'Hands-on Data Analysis' workflow (SKILL.md). 2. Boundary markers: The instructions lack specific delimiters or warnings to ignore embedded instructions within user-provided data. 3. Capability inventory: The skill utilizes Python and SQL execution environments to analyze the provided data. 4. Sanitization: No sanitization, validation, or filtering mechanisms are described for the content of processed datasets.
Recommendations
- AI detected serious security threats
- Contains 4 malicious URL(s) - DO NOT USE
Audit Metadata