writing-social-posts

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches tweet metadata from Twitter's syndication service (cdn.syndication.twimg.com) and package registry data from PyPI (pypi.org) to verify installation commands.
  • [COMMAND_EXECUTION]: Utilizes local system tools including headless Google Chrome and ffmpeg to render and process media assets for social posts.
  • [COMMAND_EXECUTION]: Executes a local Python script for image measurement using the uv tool within IMAGES.md.
  • [PROMPT_INJECTION]: The skill ingests and processes data from external social media links, creating a surface for indirect prompt injection.
  • Ingestion points: Reads external post content via curl in SKILL.md.
  • Boundary markers: No specific delimiters or safety instructions are defined for processing external text.
  • Capability inventory: Capabilities include command execution via headless Chrome and ffmpeg, and Python script execution.
  • Sanitization: External content is processed without explicit sanitization or filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 10:27 AM
Security Audit — agent-trust-hub — writing-social-posts