context-mode

Warn

Audited by Snyk on Nov 13, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (low risk: 0.10). The required workflow runs offline preflight, which reads package.json name/version and Git HEAD from a user-supplied --repo checkout that could be an externally cloned repository, so its metadata strings reach the agent's context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (medium risk: 0.30). The Claude plugin setup path instructs /plugin marketplace add mksglu/context-mode, a GitHub repository locator whose installed plugin content runs hooks, and the publisher is an individual-owned repository, so it is unofficial and its floating marketplace revision is not authenticated, giving medium; the npm and npx package names have no URL and are excluded, and the citation and metadata URLs are static or recorded origin only.

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 manages user-level hook, plugin, and settings writes and an npm postinstall that edits Claude settings and creates symlinks, but every such write is gated behind explicit approval and nothing asks for sudo, system files, new accounts, or SSH/systemd changes.

Issues (3)

W011
MEDIUM

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

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

W013
MEDIUM

Attempt to modify system services in skill instructions.

Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 13, 2026, 11:08 PM
Issues
3