seo-flow

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly pulls and loads prompt files from the public GitHub repo (github.com/AgriciDaniel/flow) via the "/seo flow sync" command and also loads prompt files on-demand per SKILL.md, meaning public, user-managed content from GitHub is ingested and used as prompts that can materially change the agent's behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill exposes a runtime "sync" operation that runs "python scripts/sync_flow.py" to pull the latest prompt files from the GitHub repo github.com/AgriciDaniel/flow — these fetched prompt files would directly control agent prompts and are required when local prompt files are missing, so the repo URL is a runtime external dependency.