x-feed-summarizer
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted content from X.com, creating a surface for indirect prompt injection where malicious tweets could attempt to manipulate the agent's summarization output.\n- Ingestion points: Tweet text and associated metadata extracted from the DOM in SKILL.md.\n- Boundary markers: Absent; content is extracted and summarized without delimiters or warnings to ignore embedded instructions.\n- Capability inventory: The skill uses agent-browser for automated navigation and script execution, along with curl for status checks.\n- Sanitization: Absent; the skill does not include logic to sanitize or escape the content of extracted tweets.\n- [COMMAND_EXECUTION]: The skill uses dynamic JavaScript execution to interact with the browser's Document Object Model (DOM) and extract information.\n- Evidence: The skill uses agent-browser --cdp 9222 eval to execute a script that parses tweet data directly from the page.
Audit Metadata