parallel-cli

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to use Bash for installing, authenticating, and running the parallel-cli tool, including management of financial balances.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download the parallel-cli utility and its dependencies from repositories managed by the parallel-web organization.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests large amounts of untrusted content from the web, creating a potential attack surface for indirect prompt injection.
  • Ingestion points: extraction, research, and findall workflows in SKILL.md and references/parallel-cli-reference.md.
  • Boundary markers: The skill advises writing results to JSON files in /tmp to avoid flooding the agent's context with raw tool output.
  • Capability inventory: The agent has Bash access for command execution and file operations, plus the ability to spawn subagents.
  • Sanitization: The instructions recommend basic noise filtering of web content but do not include rigorous sanitization of text data.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes subagent delegation for background polling of long-running research and enrichment tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 01:39 PM
Security Audit — agent-trust-hub — parallel-cli