ralph-upgrade

Warn

Audited by Snyk on Mar 9, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 1.00). The skill explicitly clones or fetches the public GitHub repository https://github.com/tolulawson/ralph-harness and instructs the agent to read and follow its UPGRADING.md and related manifest files as authoritative upgrade steps, meaning untrusted third-party content directly influences actions.

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

  • Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs fetching/cloning https://github.com/tolulawson/ralph-harness at runtime and then running code from that checkout (e.g., scripts/migrate-installed-runtime.py), so external repository content is executed and controls the upgrade flow.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 9, 2026, 10:02 AM