scan
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (CRITICAL): Automated scanner URLite detected a malicious URL blacklist entry (UR4703CF1437997CDA-0200) associated with the file
Profile.md. This indicates the skill references or contains content from known malicious entities. - [COMMAND_EXECUTION] (HIGH): The skill executes a local script via
python3 scripts/fetch_market_data.py. This grants the skill the ability to run arbitrary code on the host machine, which is a significant risk given the malicious URL detection. - [DATA_EXFILTRATION] (MEDIUM): The skill reads highly sensitive financial information from local files, including
Holdings.md(portfolios) andProfile.md(user risk profiles). The combination of sensitive data access and external data ingestion poses a data leakage risk. - [PROMPT_INJECTION] (LOW): The skill processes untrusted news items and search results, creating an indirect prompt injection surface. Evidence: 1. Ingestion points:
newsfield from script output andWebSearchresults. 2. Boundary markers: Absent. 3. Capability inventory: Shell command execution and file writing. 4. Sanitization: None identified.
Recommendations
- AI detected serious security threats
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata