The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the agent-browser CLI to perform browser interactions, including navigation, clicking, and scrolling. It also uses curl to verify the local browser's CDP connection status.
[DATA_EXFILTRATION]: The skill is designed to extract user-generated content (tweets, usernames, and timestamps) from x.com. While this is the intended purpose, it involves reading data from an external platform and returning it to the agent context.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from an external source without sanitization or clear boundary markers.
Ingestion points: Scrapes tweet data from the DOM in SKILL.md (Step 3 and Profile Scrape sections).
Boundary markers: None. The scraped text is returned as raw strings without delimiters or instructions to ignore embedded commands.
Capability inventory: The skill uses agent-browser which has capabilities for eval (JavaScript execution), navigation, and interaction.
Sanitization: No sanitization or filtering is applied to the extracted text or user fields before they are returned to the agent.

x-twitter-scraper