gardener

Fail

Audited by Snyk on Nov 17, 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 dependency-maintenance guide whose steps are disclosed and serve the stated purpose, but it tells the agent to run registry queries outside the repo so the development-engine check does not block them, and to add exact-version exceptions to the release-age policy, which weakens a supply-chain protection beyond what the task strictly needs.

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

  • Third-party content exposure detected (low risk: 0.10). The required workflow reads outsider-authored text from public npm package metadata and upstream GitHub repos/changelogs, and studies selected peer repositories, but the agent chooses which items to fetch, so an attacker can publish poison without forcing delivery.

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 is a repo-maintenance guide that edits project manifests, lockfiles, and docs in a clean clone, with no sudo, system files, or user accounts involved. It does loosen a supply-chain safeguard by adding exact-version minimumReleaseAgeExclude exceptions and steers around a development-engine check, so I'm flagging it at low-moderate severity under the recall-first instruction.

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 17, 2026, 09:54 PM
Issues
3