SubdomainEnum

Fail

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The Full Workflow instructions in SKILL.md include an example command that uses SSH to connect to a specific, hardcoded IP address (207.244.244.11) as the root user to run external database tools.
  • [REMOTE_CODE_EXECUTION]: The skill's installation guide and configuration scripts (src/configs/tools.ts) use piped shell execution to install the Bun runtime from its official domain. While the domain is well-known, this execution pattern bypasses local security checks.
  • [EXTERNAL_DOWNLOADS]: The skill downloads numerous third-party tools via go install from GitHub and fetches large datasets for cloud enumeration from external sources like kaeferjaeger.gay.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface for indirect prompt injection.
  • Ingestion points: It processes untrusted JSON output from httpx (including website titles and tech detections) and external wordlists.
  • Boundary markers: No delimiters or instructional guards are present to prevent the agent from obeying instructions embedded in this external data.
  • Capability inventory: The skill can execute shell commands via child_process.spawn and go install, and performs various file system operations.
  • Sanitization: Untrusted data is interpolated directly into markdown reports and agent summaries without validation or escaping.
  • [DATA_EXPOSURE]: The skill manages authentication secrets by reading from and writing to sensitive configuration paths, specifically ~/.claude/config/secrets.json and subfinder's provider-config.yaml.
Recommendations
  • HIGH: Downloads and executes remote code from: https://bun.sh/install - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Nov 9, 2026, 12:00 PM