autonomous-loop

Warn

Audited by Snyk on Feb 23, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). The skill's references explicitly configure MCP servers for GitHub and browser automation (references/mcp-patterns.md and the "verify_feature"/verify_ui examples) and include scripts that call a puppeteer MCP to navigate arbitrary URLs and fetch GitHub issues, meaning the agent will fetch and interpret untrusted public web/user-generated content that can influence its decisions.

MEDIUM W013: Attempt to modify system services in skill instructions.

  • Attempt to modify system services in skill instructions detected (high risk: 0.90). The prompt explicitly instructs running the agent with "--dangerously-skip-permissions" (bypassing security checks) and directs autonomous, persistent filesystem modification and init scripts, which encourages bypassing protections and potential system compromise.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 23, 2026, 02:58 AM