wp-abilities-verify
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to execute shell commands specified in an untrusted project file (AGENTS.md). This creates an indirect prompt injection surface where a malicious project could trick the agent into running harmful commands during the environment setup phase.
- [REMOTE_CODE_EXECUTION]: The skill uses the wp eval command to execute dynamically generated PHP snippets on the target WordPress instance. These snippets interpolate data derived directly from the plugin's source code, which could potentially lead to PHP code injection if the plugin contains maliciously crafted ability names or metadata.
- [SAFE]: The skill's primary functions—source code enumeration, schema validation, and permission checking—are performed using standard local tools like ripgrep and wp-cli. No patterns of data exfiltration or credential harvesting were identified.
Audit Metadata