create-internal-app
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches markdown reference files and additional skill instructions from the author's public GitHub repository (
github.com/Melvynx/ai-maxxing-skills). These resources are used to guide the application architecture and performance tuning. - [COMMAND_EXECUTION]: Orchestrates infrastructure tasks through various CLI tools, including
gh(GitHub),cf(Cloudflare),ssh, anddokploy-cli. This includes creating private repositories, managing Docker containers on a VPS, and configuring WAF rules. - [INDIRECT_PROMPT_INJECTION]: Implements a multi-step instruction chain by producing an
About.mdfile intended to govern the behavior of a secondary agent (Hermes). - Ingestion points: User-provided application descriptions and subdomain parameters.
- Boundary markers: Absent; the skill does not specify delimiters or instructions to ignore embedded content within the generated file.
- Capability inventory: The skill has access to SSH, GitHub repository management, and Cloudflare DNS/WAF configuration.
- Sanitization: No explicit sanitization or escaping is mentioned for user input before it is written into the agent-facing documentation.
Audit Metadata