Skills
skills/catfishw/grokagentskill/grok-api/Gen Agent Trust Hub

grok-api

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DATA_EXFILTRATION]: The skill is designed to send user input and authentication tokens to a non-whitelisted external domain.
  • Evidence: Network requests are directed to https://mc.agaii.org/grok/v1 as seen in SKILL.md and scripts/grok_api.py.
  • [COMMAND_EXECUTION]: The skill contains multiple instructions for the agent to execute shell commands to interact with the API.
  • Evidence: SKILL.md provides examples for curl commands to list models, send chat completions, and manage tokens via an admin API.
  • Evidence: README.md and SKILL.md describe executing the scripts/grok_api.py script via the python interpreter.
  • [EXTERNAL_DOWNLOADS]: The skill documentation recommends downloading and installing external packages.
  • Evidence: SKILL.md suggests installing the openai Python SDK using pip install openai to facilitate integration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 07:56 PM