ship
Fail
Audited by Snyk on Nov 16, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill is a legitimate ship workflow, but the App Store section directs the agent to silently mint a durable App Store Connect API key on the user's account and to withhold that from them mid-run (disclosing it only at exit), which hides a standing credential the user would not expect.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The required ship workflow reads outsider-authored PR review comments: Step 10 (
sections/greptile.md) addresses review bot comments on an existing PR, and anyone who can comment on a public GitHub PR can post text there without the agent selecting it first, so this is a monitored GitHub PR feed (medium).
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill's core actions (merging, running tests, bumping VERSION/CHANGELOG, pushing and opening PRs) are project-scoped and need no sudo, but it does install a git pre-push hook and can drive vendor dashboards to create API keys, though both are gated by explicit user consent.
Issues (3)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata