parallel-data-enrichment

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: Fetches and executes an installation script from the vendor's domain (parallel.ai) via a piped shell command (curl | bash). This is a standard installation method for the service and originates from the author's infrastructure.
  • [EXTERNAL_DOWNLOADS]: Offers an alternative setup method via pipx to install the parallel-web-tools package from the Python Package Index.
  • [COMMAND_EXECUTION]: Utilizes the parallel-cli tool through the bash environment to run enrichment tasks, poll for results, and manage configuration.
  • [PROMPT_INJECTION]: Identifies a surface for indirect prompt injection as the skill processes external data from CSV files and JSON strings provided by the user.
  • Ingestion points: Reads company and product information from local CSV files and inline arguments.
  • Boundary markers: The instructions do not define specific delimiters or warnings for the agent to ignore instructions embedded in the processed data.
  • Capability inventory: The skill uses bash to execute CLI tools with arguments directly derived from processed inputs.
  • Sanitization: No explicit validation or escaping of input data is required by the instructions before it is passed to shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 04:26 PM