counterparty-guard

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script scripts/fetch_counterparty.py to automate the retrieval of tax and financial data. This script uses the Python standard library and contains defensive parsing logic to handle network responses safely.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with well-known business data services including checko.ru, list-org.com, saby.ru, and rusprofile.ru, as well as official Russian government domains such as nalog.ru and nalog.gov.ru. These resources are standard for the skill's stated purpose of counterparty due diligence.
  • [PROMPT_INJECTION]: The skill processes external data which could theoretically contain malicious prompts (Indirect Prompt Injection). However, the skill incorporates a 'Safety-floor' section that explicitly instructs the agent to treat all scraped content as data, not as instructions, effectively mitigating this risk surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 06:56 PM
Security Audit — agent-trust-hub — counterparty-guard