pp-claude-agent-sdk-python-docs

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses the CLI’s “pages”/docs-intelligence commands (e.g., pages, context, recipe, verify) which fetch public Claude Code docs over HTTPS and then ingest the fetched page text/snippets into the agent’s LLM context, creating an outsider free-text injection surface.

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

• Potentially malicious external URL detected (high risk: 0.70). The CLI explicitly performs live retrieval of the "public Claude Code docs over HTTPS" at runtime (the SKILL.md says the pages/docs intelligence commands "read the public Claude Code docs over HTTPS"), and that remote content is used to build agent context and thus directly influences agent prompts.