parallel-deep-research

Fail

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill contains a command to download and execute a script directly from the internet via curl -fsSL https://parallel.ai/install.sh | bash. This pattern is highly insecure as it executes unverified code with shell privileges and was flagged by automated security scans.
  • [COMMAND_EXECUTION]: The skill uses parallel-cli with variables $ARGUMENTS, $RUN_ID, and $FILENAME interpolated directly into shell commands. This presents a risk of command injection if the inputs are not properly sanitized by the underlying system.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of third-party software and scripts from parallel.ai and PyPI via pipx install parallel-web-tools.
  • [CREDENTIALS_UNSAFE]: The skill provides instructions for the user to manually enter and export an API key in the shell environment using export PARALLEL_API_KEY, which is a common but insecure method for handling sensitive credentials that can lead to exposure in shell history or environment logs.
Recommendations
  • HIGH: Downloads and executes remote code from: https://parallel.ai/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Mar 2, 2026, 11:26 PM