ghost-scan-secrets
Warn
Audited by Snyk on Feb 20, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The skill explicitly downloads and executes third-party code from GitHub (agents/init/agent.md uses curl https://raw.githubusercontent.com/... | bash and README.md describes auto-downloading GitHub releases), so it ingests and runs untrusted public content that can alter agent behavior.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). The init agent runs a runtime command that curls and pipes https://raw.githubusercontent.com/ghostsecurity/poltergeist/main/scripts/install.sh to bash to install the required poltergeist binary, which fetches and executes remote code and thus directly controls runtime behavior.
Audit Metadata