ai-social-media-content

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs users to download and install the infsh CLI tool and refers to documentation hosted on GitHub at raw.githubusercontent.com/inference-sh/skills/.- [EXTERNAL_DOWNLOADS]: The skill suggests incorporating additional functionality by adding related skills from the inference-sh/skills repository using the npx skills add command.- [COMMAND_EXECUTION]: Extensive use of Bash commands is provided for interacting with the infsh CLI, including account authentication (infsh login) and executing inference tasks for various AI models.- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection where user-supplied descriptions and scripts are processed by AI models.
  • Ingestion points: User-provided text prompts for image, video, and text generation tasks in SKILL.md.
  • Boundary markers: Shell commands utilize structured JSON input objects (--input '{...}') to encapsulate prompts.
  • Capability inventory: The skill uses the Bash(infsh *) tool to perform remote AI inference and local file operations (e.g., writing results to .json files).
  • Sanitization: No client-side content sanitization is implemented; the skill relies on the safety filters of the underlying inference service provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 11:26 PM