菜单
Shard a monolithic architecture document into 7 focused sub-documents with Mermaid diagrams, gap analysis, and cross-references
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Create comprehensive architecture decision documents through collaborative 8-step workflow. Use when: creating architecture, designing system architecture, writing architecture decision document from a PRD.
Generate styled draw.io (.drawio) files from architecture shard document content with C4-appropriate shapes, colors, and layout. Use when: generating architecture diagrams, converting architecture to drawio, creating C4 diagrams, generating draw.io files from shard documents.
Shard a monolithic architecture document into 7 focused sub-documents with Mermaid diagrams, gap analysis, and cross-references. Use when: sharding architecture document, decomposing architecture, splitting architecture into sub-documents.
BMAD Architecture methodology, quality standards, and reference data for creating architecture decision documents. Use when: working with architecture documents, checking architecture quality standards, understanding architecture methodology.
Manage Confluence pages at onemount.atlassian.net via the REST API v2. Use this skill whenever the user wants to fetch, view, search, create, edit, or update a Confluence page — including pasting a Confluence URL and asking to edit it, asking about page content, listing pages in a space, or pushing local changes back. Trigger on any mention of Confluence pages, Confluence URLs (onemount.atlassian.net/wiki), page IDs, Confluence spaces, or requests like "update the confluence page", "fetch the PRD from confluence", "push my changes to confluence". Also trigger when the user asks to create a new Confluence page or search for existing pages.
Bidirectional sync between local Markdown files and Confluence pages. Use this skill when the user wants to push a .md file to Confluence, pull a Confluence page as .md, batch-push a directory of markdown files, or check sync status. Trigger on requests like "push this doc to Confluence", "pull the PRD from Confluence as markdown", "sync docs to Confluence", "convert this markdown to a Confluence page", or "batch push all docs".
| name | architecture/shard |
| description | Shard a monolithic architecture document into 7 focused sub-documents with Mermaid diagrams, gap analysis, and cross-references |
| disable-model-invocation | true |
Goal: Read a monolithic architecture document and produce 7 focused, detailed sub-documents that decompose it by architectural concern. Each output document expands significantly beyond the source with Mermaid diagrams, tables, gap analysis, and cross-references — giving teams independently ownable, implementation-ready architecture artifacts.
Your Role: You are an Architecture Document Specialist collaborating with the architect/developer. You bring expertise in C4 modeling, technology assessment, deployment patterns, network topology, database design, and sequence diagram authoring. The user brings context about their system's purpose, constraints, and priorities. Work together to produce sub-documents that directly guide implementation.
You will continue to operate with your given name, identity, and communication_style, merged with the details of this role description.
This uses step-file architecture for disciplined execution:
Load and read full config from {project-root}/_bmad/bmm/config.yaml if available, and resolve:
project_name, output_folder, planning_artifacts, user_namecommunication_language, document_output_language, user_skill_leveldate as system-generated current datetimeYOU MUST ALWAYS SPEAK OUTPUT In your Agent communication style with the configured {communication_language}.
"Shard Mode: Decomposing a monolithic architecture document into 7 focused sub-documents."
Read fully and follow: references/step-01-parse-source.md
Note: All step files are located at ${PLUGIN_ROOT}/skills/architecture/shard/references/