wp-abilities-audit

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow reads and parses the target plugin’s PHP source (via the provided checkout path), extracting free-text fields like Notes and Surprises and proposed ability metadata, so if the audited plugin is supplied by an outsider (non-operator) those outsider-authored comments/docblocks and code are ingested into the agent’s LLM context for summarization.