The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs users to install a third-party plugin from a public marketplace via the command /plugin marketplace add mvanhorn/last30days-skill. This introduces a supply chain risk, as the plugin is hosted under an unverified user account. It also recommends installing yt-dlp for media handling.
[COMMAND_EXECUTION]: The documentation encourages the use of local automation tools such as xhs_ai_publisher (Playwright-based) and Autoxhs for automated content publishing. These involve running scripts that interact with the host system and external web interfaces.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of fetching data from untrusted external sources.
Ingestion points: Social media content fetched from Reddit, X (Twitter), TikTok, Instagram, and YouTube via the /last30days tool.
Boundary markers: None documented to prevent the agent from following instructions embedded in the scraped content.
Capability inventory: Generates content strategies, titles, and image prompts based on the fetched data.
Sanitization: No evidence of sanitization or filtering of external input before it is processed by the LLM.

xiaohongshu-ops