wordpress-setup

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs creating and using WordPress application passwords and shows embedding them verbatim in curl commands and in a plaintext .dev.vars file, so the agent would need to include actual secret values in generated outputs (commands/files), creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and interprets data from arbitrary WordPress sites (e.g., WP-CLI SSH commands like "wp --ssh=user@host/path option get siteurl" and REST API calls such as "curl https://example.com/wp-json/wp/v2/posts...") as part of its required workflow in SKILL.md, so untrusted third-party content can influence verification, configuration, and subsequent actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's setup instructs fetching and installing WP-CLI at runtime via curl from https://raw.githubusercontent.com/wp-cli/builds/gh-pages/phar/wp-cli.phar, which downloads executable remote code the skill relies on to run WP-CLI commands.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The prompt explicitly instructs using sudo to install wp-cli into /usr/local/bin (modifying system directories), suggests disabling SSH strict host key checking and using --allow-root, which directs the agent to obtain/use elevated privileges and bypass security mechanisms on the host.