Skills
Navigate
skills/glebis/claude-skills/sorted/Gen Agent Trust Hub

sorted

Fail

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill contains hardcoded sensitive financial and personal identifiers.
  • Evidence: Hardcoded VAT-ID (DE369692682), Tax Number (2337201265), and IBAN (DE38 1001 0178 8157 1777 30) are included in the 'Gleb's Sorted Account Details' section.
  • Evidence: The skill instructs the agent to read sensitive API credentials (api_id and api_hash) and session data from local files: ~/.telegram_dl/config.json and ~/.telegram_dl/user.
  • [COMMAND_EXECUTION]: The skill relies on the execution of external CLI tools and scripts to interact with the browser and network.
  • Evidence: Requirement for the agent-browser CLI tool and execution of shell commands for session setup.
  • [DYNAMIC_EXECUTION]: The skill provides a Python script intended to be executed by the agent at runtime for file delivery.
  • Evidence: The 'Sending Invoice After Download' section contains a Python code block using the telethon library for Telegram automation.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 19, 2026, 04:42 AM