菜单
Understand why a file or function exists — git history, decision context, team annotations. Use when the user says "/sprang-why", "why does this exist", "who wrote this", "history of this file", or "what decisions led to this".
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
Analyze a codebase to produce a rich semantic knowledge graph — file summaries, architecture layers, guided tour, domain map, risk scores. Use when the user says "/sprang-analyze", "analyze the codebase", "full analysis", or "run sprang-analyze".
Analyze a codebase to produce a rich semantic knowledge graph — file summaries, architecture layers, guided tour, domain map, risk scores. Use when the user says "/sprang-analyze", "analyze the codebase", "full analysis", or "run sprang-analyze".
Build or refresh the Sprang knowledge graph for this workspace. Use when the user says "/sprang", "build the knowledge graph", "scan the codebase", "index this project", or "run sprang".
Build or refresh the Sprang knowledge graph for this workspace. Use when the user says "/sprang", "build the knowledge graph", "scan the codebase", "index this project", or "run sprang".
Guided architecture tour for new team members — adapts to junior, senior, or PM persona. Use when the user says "/sprang-onboard", "onboard me", "give me a tour", "I'm new to this codebase", or "walk me through the architecture".
Browse and write team annotations attached to graph nodes — institutional knowledge, decision history, ownership. Use when the user says "/sprang-team", "add annotation", "team knowledge", "who owns this", or "document this decision".
| name | sprang-why |
| description | Understand why a file or function exists — git history, decision context, team annotations. Use when the user says "/sprang-why", "why does this exist", "who wrote this", "history of this file", or "what decisions led to this". |
sprang_query with $ARGUMENTS to find the matching graph node ID (file path, function name, class name).sprang_why with the resolved node_id to retrieve: commit history, primary authors, rationale snippets, PR references, and any team annotation content.sprang_node with the same node_id to get enriched structural context:
layer (name + id) and layer_mate_count — where it fits in the architecturein_degree / out_degree — how coupled it ishas_annotation / annotation_path — whether team knowledge existshas_annotation is true, display the full annotation content prominently.decision_context.change_frequency >= 10 (commits in 90 days), flag as 🔴 frequently-churning — extra care required.risk_score > 0.7).