compozy

Warn

Audited by Snyk on Jul 17, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.85). Yes—during the required review/remediation workflow, compozy fetch-reviews pulls outsider-authored free-form review-body text from an external provider (e.g., CodeRabbit) and writes it into reviews-NNN/issue_*.md, which is then read by the LLM in compozy fix-reviews/cy-fix-reviews as prompt context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 1.00). The docs instruct installing an extension from the GitHub repo reference "compozy/compozy" via commands like "compozy ext install --yes compozy/compozy --remote github --ref --subdir extensions/cy-idea-factory", which fetches remote extension code that is then executed at runtime (executable extensions can perform prompt decoration, plan injection, and agent session modification) and is required to use the cy-idea-factory skill.

Issues (2)

W011
MEDIUM

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

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 17, 2026, 06:51 PM
Issues
2