workos-authkit-nextjs

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

• Third-party content exposure detected (high risk: 1.00). The skill's Step 1 explicitly requires a blocking WebFetch of the public GitHub README (https://github.com/workos/authkit-nextjs/blob/main/README.md) and instructs the agent to treat that external README as the source of truth, meaning untrusted third-party content is read and can change subsequent tool use and decisions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly performs a blocking WebFetch of the README at https://github.com/workos/authkit-nextjs/blob/main/README.md and states the README is the "source of truth" that must be followed, so this runtime-fetched content directly controls the skill's instructions and is a required dependency.