skills/anbeime/skill/baoyu-post-to-x/Gen Agent Trust Hub

baoyu-post-to-x

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various system tools and utilities across different platforms to facilitate automation and clipboard management.
  • Evidence in scripts/copy-to-clipboard.ts: Runs swift on macOS, xclip or wl-copy on Linux, and powershell.exe on Windows.
  • Evidence in scripts/paste-from-clipboard.ts: Runs osascript on macOS, xdotool or ydotool on Linux, and powershell.exe on Windows.
  • Evidence in scripts/x-utils.ts: Executes local TypeScript scripts using npx -y bun.
  • [EXTERNAL_DOWNLOADS]: The skill fetches remote assets from URLs provided within Markdown files during the article processing workflow.
  • Evidence in scripts/md-to-html.ts: The downloadFile function uses standard HTTP/HTTPS modules to download files (intended as images) from arbitrary remote locations to a local temporary directory.
  • [DYNAMIC_EXECUTION]: The skill generates and executes platform-specific scripts and code at runtime to perform system-level operations.
  • Evidence in scripts/copy-to-clipboard.ts: Dynamically generates and executes a Swift script to handle rich text and image clipboard data on macOS.
  • Evidence in scripts/paste-from-clipboard.ts: Constructs and executes AppleScript code via osascript to simulate keyboard paste events on macOS.
  • Evidence in scripts/x-article.ts, scripts/x-browser.ts, and scripts/x-video.ts: Uses the Chrome DevTools Protocol Runtime.evaluate command to execute JavaScript directly within the browser session.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts on data from external sources, such as user-provided Markdown files and text, which can influence its execution flow.
  • Ingestion points: Processes content from Markdown files in scripts/x-article.ts and scripts/md-to-html.ts, and accepts user-provided text for social media posts.
  • Boundary markers: There are no explicit markers or instructions defined in the scripts to isolate or ignore potentially malicious instructions embedded in the processed content.
  • Capability inventory: The skill possesses capabilities for network access, file system modifications, and system command execution.
  • Sanitization: Employs basic HTML escaping in scripts/md-to-html.ts but does not perform comprehensive validation of the external content's structure or intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 10:52 AM