project-build-from-scratch

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a structured design and build process for new software applications. It emphasizes user control through a mandatory architecture selection step before any project scaffolding occurs.\n- [COMMAND_EXECUTION]: The skill utilizes standard project initialization commands such as dotnet new, ng new, and ionic start. These are common industry-standard tools for creating project structures and their use is well-defined within the skill's intended scope.\n- [PROMPT_INJECTION]: The skill ingests user-provided application specifications to generate design options. The risk of malicious instructions in the specification is mitigated by the "Design-then-Pick" workflow, which ensures the user reviews and approves the technical stack and architecture before implementation begins.\n
  • Ingestion points: User specification provided in the session conversation (SKILL.md).\n
  • Boundary markers: Explicit requirement to ask the user for clarification on spec gaps and grounding in specified house skills.\n
  • Capability inventory: Project scaffolding via CLI commands (dotnet, ng, ionic) and file system writes for code implementation.\n
  • Sanitization: Mandatory user approval gate ("THE PICK") before scaffolding or code generation starts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 07:50 AM
Security Audit — agent-trust-hub — project-build-from-scratch