skills/membranedev/application-skills/cloudback/Snyk

cloudback

Warn

Audited by Snyk on Apr 30, 2026

Risk Level: MEDIUM

Full Analysis

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

Third-party content exposure detected (high risk: 0.80). The SKILL.md directs the agent to use the Membrane CLI to run actions against Cloudback (and linked SaaS like Jira, Trello, GitHub) with results returned in the action "output" field, which means the agent will ingest and act on arbitrary user-generated/third-party data from those services.

Issues (1)

W011

MEDIUM

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

Audit Metadata

Risk Level

MEDIUM

Analyzed

Apr 30, 2026, 07:59 AM

Issues

1

Security Audit — snyk — cloudback