suparank/setup

Warn

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: MEDIUMCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill guides the user to provide highly sensitive secrets, including WordPress application passwords, Ghost Admin API keys, and fal.ai API keys.
  • [DATA_EXFILTRATION]: The skill accesses sensitive file paths outside the project directory, specifically reading from ~/.suparank/credentials.json for migration and writing to ~/.claude/suparank-credentials.json.
  • [EXTERNAL_DOWNLOADS]: The skill recommends that the user execute a command that downloads and runs code from an unverified remote source: https://seo-mcp.cnych.workers.dev/sse.
  • [COMMAND_EXECUTION]: The skill automatically executes a shell command to initialize directories: mkdir -p .claude/suparank-content.
  • [PROMPT_INJECTION]: The skill has a significant data ingestion surface (Category 8) as it takes untrusted user input for numerous configuration fields and writes them directly into a JSON file without sanitization.
  • Ingestion points: User-provided strings for site name, description, brand voice, audience, etc., in SKILL.md.
  • Boundary markers: None used when collecting or writing the data.
  • Capability inventory: Subprocess calls for mkdir, file read/write operations for configuration and credential files.
  • Sanitization: None; input is interpolated directly into the JSON structure.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 28, 2026, 09:35 PM
Security Audit — agent-trust-hub — suparank/setup