wp-project-triage

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted files from a repository to detect project structure and metadata.
  • Ingestion points: The scripts/detect_wp_project.mjs script reads file contents (e.g., .php, .js, .json, .css) from the target directory using fs.readFileSync to identify specific signatures and configuration constants.
  • Boundary markers: None; the script reads raw file contents into memory for analysis.
  • Capability inventory: The skill is restricted to read-only filesystem access and standard output (stdout). It does not contain capabilities for writing files, executing external commands, or making network requests.
  • Sanitization: Content extracted from files is processed using regular expressions and string matching, and then encapsulated in a structured JSON object for output.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:51 PM
Security Audit — agent-trust-hub — wp-project-triage