wordpress-setup

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the WP-CLI tool (wp-cli.phar) from the official WP-CLI repository on GitHub. This is a well-known and established tool in the WordPress ecosystem.
  • [PRIVILEGE_ESCALATION]: Instructions include using sudo to move the WP-CLI binary to /usr/local/bin/ during the installation process, which is a standard administrative task for CLI tool installation. It also mentions the --allow-root flag for WP-CLI operations where hosting environments require it.
  • [COMMAND_EXECUTION]: The skill executes various bash commands including wp, curl, and ssh to retrieve site metadata, verify connectivity, and manage configuration files. These commands are integral to the stated purpose of WordPress site management.
  • [INDIRECT_PROMPT_INJECTION]: The skill has the attack surface for indirect prompt injection as it retrieves data from external WordPress installations (such as post titles and plugin lists) and incorporates it into the agent's context.
  • Ingestion points: wp post list, curl .../wp-json/v2/posts (SKILL.md).
  • Boundary markers: None explicitly defined in the provided commands.
  • Capability inventory: File system writes (wp-cli.yml, .dev.vars, wordpress.config.json), network operations (curl, ssh), and command execution.
  • Sanitization: Standard CLI output processing is used; no explicit input sanitization is mentioned for remote content.
  • [CREDENTIALS_SAFE]: The skill provides instructions for storing authentication data. It explicitly recommends adding .dev.vars to .gitignore to prevent accidental credential leakage and suggests using dedicated secret managers for cross-project use.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:26 PM
Security Audit — agent-trust-hub — wordpress-setup