wp-plugin-development
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The
detect_plugins.mjsscript reads metadata (such as Plugin Name, Description, and Author) directly from PHP files in the repository to identify active plugins. This metadata is presented to the agent, which could allow maliciously crafted files to influence the agent's behavior via the ingested strings. - Ingestion points: The
scripts/detect_plugins.mjsscript reads the first 128KB of every PHP file found in the project root to parse WordPress plugin headers. - Boundary markers: There are no explicit boundary markers or instructions to the agent to treat this specific metadata as untrusted data once it is aggregated into the discovery report.
- Capability inventory: The skill defines procedures for code modification, database migration, and security implementation (nonces, capability checks) across its reference files.
- Sanitization: The script extracts header strings using regular expressions and provides them in a JSON report without sanitizing the content, which might contain natural language instructions.
- [COMMAND_EXECUTION]: The skill uses local Node.js scripts to perform repository triage and plugin detection. The discovery script (
detect_plugins.mjs) is implemented with safety guardrails, including a maximum file count (5000) and recursion depth (10) to prevent resource exhaustion during filesystem operations.
Audit Metadata