pp-wordpress

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs agents to replay playbook steps with Playbook.slots_resolved substitutions (which may include a "token" / live token) and also references setting WORDPRESS_APP_PASSWORD, meaning an LLM replaying or emitting the substituted commands could be required to output secret values verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s “Agent Mode” and “Automatic learning” describe that the CLI expands --agent and then reads recall outputs (including query, candidates, playbook, notes, and results) derived from the agent/user question, which at runtime is free text submitted by the user.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's prerequisites require running remote installers that fetch and execute code (npm and Go installs) — e.g. "npx -y @mvanhorn/printing-press-library install wordpress --cli-only" and "go install github.com/mvanhorn/printing-press-library/library/developer-tools/wordpress/cmd/wordpress-pp-cli@latest" — which are runtime installation steps that the skill depends on.