Skills
skills/steveonead/agent-skills/ito-search/Gen Agent Trust Hub

ito-search

Fail

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: CRITICAL
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses tools like the GitHub CLI (gh) and Model Context Protocol (MCP) integrations to query external repositories and documentation. This execution is limited to data retrieval as part of the skill's search functions.
  • [EXTERNAL_DOWNLOADS]: Fetches data from external websites using tools like exa and built-in web fetchers. The skill mitigates risks associated with untrusted content by applying a blacklist filtering step before presenting information to the user.
  • [PROMPT_INJECTION]: The skill includes instructions to strictly adhere to a blacklist defined in references/source-filter.md. This mechanism helps prevent the inclusion of instructions or data from malicious sites like isolution.pro, which was identified by automated scans as untrusted.
Recommendations
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Apr 29, 2026, 02:21 AM