The Agent Skills Directory

[COMMAND_EXECUTION]: The fetch_direct.py script utilizes subprocess.run() to invoke the main scraping tool. It correctly passes arguments as a list rather than a single string, which prevents shell injection vulnerabilities when handling the user-provided URL.
[EXTERNAL_DOWNLOADS]: The skill relies on the Playwright automation library, which typically downloads browser binaries (e.g., Chromium) to the system. This is standard behavior for the tool's primary purpose.
[PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection because it fetches and processes arbitrary text from X (Twitter). If a tweet contains malicious instructions, the agent may inadvertently follow them when analyzing the fetched content.
Ingestion points: Untrusted tweet content is retrieved from X.com in x_tweet_fetcher.py via the XTweetFetcher class.
Boundary markers: No delimiters or protective markers are used to encapsulate the external text when presented to the agent.
Capability inventory: The skill has the ability to write to the filesystem and execute local scripts via subprocess.
Sanitization: There is no evidence of filtering or sanitizing the retrieved tweet content before processing.

x-tweet-fetcher