vp-cspell

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly requires fetching cspell syntax/config at runtime from Context7 (/streetsidesoftware/cspell) and falls back to the public cspell.org website, so the agent will ingest and act on content from public third‑party sources (a public repo/site) that can materially influence config/init decisions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly fetches cspell configuration/schema from Context7 at runtime ("/streetsidesoftware/cspell") and falls back to https://cspell.org/configuration/, which means external content would be injected into the agent's decision flow and could directly control prompts/instructions.