om-auto-merge-buddy

Fail

Audited by Snyk on Nov 21, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The skill's merge-bypass and human-gate-skipping behaviors are openly documented and conditioned on fresh evidence and operator-applied labels, but the host-protection bypass could plausibly weaken repository protections beyond what routine automerge needs, so under recall-first I rate it medium rather than no.

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

  • Third-party content exposure detected (medium risk: 0.30). The required workflow reads GitHub PR state, including PR bodies, diffs, conversation and review comments, and failed CI logs, for any queued PR, including other people's PRs that an outsider can open or comment on, so outsider-authored text reaches the agent without it selecting a specific item.

MEDIUM W013: Attempt to modify system services in skill instructions.

  • Attempt to modify system services in skill instructions detected (low risk: 0.10). The skill performs operational changes, including launching a persistent local dashboard service and mutating scheduler state, merge and tracker state, but no explicit irreversible operational loss is established, so it stays at low.

Issues (3)

E004
CRITICAL

Prompt injection detected in skill instructions.

W011
MEDIUM

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

W013
MEDIUM

Attempt to modify system services in skill instructions.

Audit Metadata
Risk Level
CRITICAL
Analyzed
Nov 21, 2026, 04:26 PM
Issues
3