x-mcp

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the @xdevplatform/xurl package via npm to handle X API interactions.
  • [COMMAND_EXECUTION]: Employs a Python driver (scripts/oauth_driver.py) that uses subprocess.Popen to manage an interactive headless OAuth process.
  • [COMMAND_EXECUTION]: Modifies the environment's setup.sh to ensure the xurl tool is installed and its credential store is symlinked to persistent storage on every boot to maintain the OAuth session across restarts.
  • [PROMPT_INJECTION]: Creates a surface for indirect prompt injection by reading external content (tweets) through MCP tools.
  • Ingestion points: MCP read tools such as search_posts_all and get_users_timeline.
  • Capability inventory: REST write actions (post, delete, reply, like, retweet, follow, DM) via the xurl CLI.
  • Boundary markers: None specified.
  • Sanitization: None described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 07:44 PM
Security Audit — agent-trust-hub — x-mcp