wp-guard
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external WordPress code, which creates a surface for indirect prompt injection. Malicious instructions could be embedded in the PHP or JS files being reviewed to attempt to influence the agent's behavior.
- Ingestion points: The skill processes "generated or changed WordPress code" and "target files" during its review passes, as defined in
SKILL.md. - Boundary markers: The skill does not define specific string delimiters for the code it reads, but it implements a "Mandatory confirmation gate" that requires user authorization before the agent performs any action based on its review.
- Capability inventory: The skill instructions do not contain subprocess calls, network exfiltration, or file-writing capabilities; it primarily acts as a reviewer and advisor.
- Sanitization: There is no instruction to sanitize the input code before analysis, relying instead on the agent's internal logic and the stated confirmation rules.
Audit Metadata