pkl

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

• Third-party content exposure detected (high risk: 0.85). The skill’s required workflow is to use WebFetch(...) to fetch Apple’s PKL documentation pages at runtime, and that fetched public web content is read as free-form text and fed into the agent’s LLM context (public web content / arbitrary fetched pages).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly mandates runtime WebFetch of PKL docs (e.g. https://pkl-lang.org/main/current/language-reference/index.html, https://pkl-lang.org/main/current/language-tutorial/01_basic_config.html, https://pkl-lang.org/main/current/pkl-cli/index.html), which the agent must fetch and follow before proceeding, so the externally fetched content directly controls the agent's instructions and is a required dependency.