xl1-build
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and links to documentation, architectural principles, and checklists hosted on the official GitHub repositories for XYOracleNetwork and Ariestools.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest exploratory or vague user requests to generate a concrete dApp specification.
- Ingestion points: User input collected during Phase 1 (Archetype identification), Phase 2 (Detail refinement), and Phase 3 (Environment capture).
- Boundary markers: The skill uses a fixed prompt template and a structured Markdown format for the PRD.md file to delimit the generated specification.
- Capability inventory: The skill has the ability to write a PRD.md file to the local filesystem and pass the generated prompt context to the
xl1-scaffoldskill for project bootstrapping. - Sanitization: The process includes a mandatory human-in-the-loop step where the agent must present the synthesized prompt for user confirmation or editing before writing any files or triggering the handoff to the scaffold tool.
Audit Metadata