twitter-x-profile-lead-scraper
Fail
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to install the 'BrowserAct CLI' dependency automatically and without requesting user confirmation if it is found to be missing. This removes the human-in-the-loop security check for software installation on the host system.
- [COMMAND_EXECUTION]: The instructions explicitly tell the agent to 'not request separate approval for dependency installation' when obtaining the required CLI tool, promoting unverified command execution.
- [EXTERNAL_DOWNLOADS]: The skill references an external site for dependency acquisition and automatic installation instructions.
- [PROMPT_INJECTION]: The skill processes untrusted user-generated content from social media platforms, posing a risk of indirect prompt injection.
- Ingestion points: Target profile URLs, usernames, hashtags, post URLs, and bio text from Twitter/X (SKILL.md).
- Boundary markers: Absent; there are no instructions to differentiate between agent commands and scraped profile data.
- Capability inventory: The agent can invoke the browser-act skill to interact with websites and export results to local files like CSV.
- Sanitization: Absent; no filtering or validation logic is defined for the external data being processed.
Recommendations
- AI detected serious security threats
Audit Metadata