committee
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: CRITICALCOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- COMMAND_EXECUTION (MEDIUM): The skill executes a shell command to run
python3 scripts/fetch_market_data.py. Executing local scripts allows the agent to perform actions on the host system, which could be malicious if the script is unverified. - DATA_EXFILTRATION (MEDIUM): The skill reads the
Profile.mdfile to determine user risk preferences. This file was flagged as 'Blacklist' by a security scanner, suggesting it contains harmful links or content intended for data exfiltration. - EXTERNAL_DOWNLOADS (LOW): The
fetch_market_data.pyscript is used to retrieve technical market data from external sources. This dependency on external data providers introduces a potential risk of ingesting malicious content. - PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection because it ingests untrusted market data and processes it across multiple model rounds without explicit sanitization or boundary markers. Ingestion Point:
fetch_market_data.py. Capabilities: Bash execution and file writing.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata