skills/mblode/ghostwriter/train-ghostwriter/Snyk

train-ghostwriter

Warn

Audited by Snyk on Jul 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: 0.85). OUTSIDER free text comes from the user-provided normalized JSONL corpus/heldout files (e.g., corpus/train.jsonl and corpus/heldout.jsonl) into the agent LLM prompt via scripts/run-agent.ts (spawnCaptured(..., stdin: prompt)), and those text fields are not authored by the skill itself.

Issues (1)

W011

MEDIUM

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

Audit Metadata

Risk Level

MEDIUM

Analyzed

Jul 23, 2026, 11:00 PM

Issues

1

Security Audit — snyk — train-ghostwriter