twitter-automation

Pass

Audited by Gen Agent Trust Hub on Nov 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed to interface with the inference.sh platform through the belt CLI tool. All referenced domains (inference.sh) and GitHub repositories (github.com/inference-sh) are consistent with the service's official infrastructure.
  • [SAFE]: Tool access is explicitly restricted in the frontmatter to Bash(belt *), implementing a least-privilege execution environment that prevents the execution of unauthorized shell commands.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text for tweet creation and direct messages. This creates a surface for indirect prompt injection where content being processed could contain instructions for the agent. However, this is the core intended functionality of a social media automation skill and is handled via structured JSON inputs to the CLI, which provides some degree of boundary separation.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 18, 2026, 08:23 PM