skills/taxueseek/argo/ego-search/Gen Agent Trust Hub

ego-search

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the ego-browser CLI tool via subprocess.run to handle browser automation and runtime management (scripts/ego_search.py, scripts/runtime.py).
  • [DYNAMIC_EXECUTION]: JavaScript code is dynamically generated from templates and executed within the browser instance to extract structure from search results and web pages (scripts/ego_search.py).
  • [EXTERNAL_DOWNLOADS]: The core browser runtime is downloaded from the official vendor website at https://lite.ego.app/ as part of the installation process (references/install.md).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted web content which could contain malicious instructions. It mitigates this by using structured extraction logic and recommending evidence-based verification of results (SKILL.md).
  • [DATA_EXFILTRATION]: The skill is designed to access sensitive authenticated browser data (login states). Security is maintained through a mandatory 'pro-mode' user activation gate and strict isolation that prevents authenticated content from entering public search caches (scripts/ego_search.py, SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 06:54 AM