What problem does it solve?
Writing large design proposals for Backend.AI often leads to inconsistent structure, missing decisions, and documents too long for reviewers or AI agents to navigate. This Skill standardizes how BEPs (Backend.AI Enhancement Proposals) are created, segmented, and maintained.
Core Features & Use Cases
- End-to-End BEP Workflow: Covers number reservation, JIRA issue creation, branch naming, metadata frontmatter, section-by-section writing guidance, and PR submission.
- Document Segmentation: Splits BEPs over 500 lines into a concise main document plus per-component sub-documents with a document index.
- AI Context Blocks: Defines
context-for-ai comment blocks so AI agents can quickly parse scope, constraints, and decisions without reading entire documents.
- Use Case: An engineer planning a scheduler change reserves BEP-1046, writes a segmented proposal with a Decision Log, and later assigns an AI agent to implement Phase 1 by reading only the relevant sub-documents.
Quick Start
Ask the AI to create a new BEP for your proposed feature following the bep-guide workflow, starting by reserving a number in the proposals registry.