dsh-kantu
Evidence-driven architecture reconnaissance for DeepSeek Harness
AI Analysis
核心用途是扫描工作区并生成系统架构文档。适合需要对代码库进行深度依赖分析、索引构建和架构可视化呈现的开发团队。
Install
$
npx -p @deepseek-ai/dsh dsh plugin --profile web add github:lvxinrong/KantuREADME
Read the full README ↗Configuration
| Option | Default | Purpose |
|---|---|---|
workspaceRoot | Current DSH session workspace | Optional scan-root override; relative values resolve from the session workspace |
outputDirectory | kantu_docs | Artifact directory inside the workspace |
discoveryMaxDepth | 3 | Maximum recursive depth for Git-root discovery |
codebaseMemoryServerName | codebase_memory_mcp | DSH namespace for codebase-memory MCP tools |
indexMode | moderate | fast, moderate, or full mode for new or refreshed indexes |
evidenceProvider | spawn | DSH subagent provider for isolated read-only evidence workers |
systemConcurrency | 4 | Maximum concurrent index and evidence tasks |
registerCommand | true | Register the optional /kantu command |
registerSystemScanTool | true | Register kantu_scan_system |
registerStatusTool | true | Register kantu_status |
The Kantu bundle also mounts the official DeepSeek Harness @deepseek-ai/dsh-mcp-client and starts codebase-memory-mcp over stdio. Before starting DSH, make sure that executable is available on PATH:
command -v codebase-memory-mcp
After installing the plugin, verify that both configuration layers are present:
dsh --profile web --dump-config | rg -n -C 6 "kantu-codebase-memory|codebase_memory_mcp|dsh-kantu"
For local development, run pnpm install followed by pnpm check. Kantu is not yet published to npm, so the public install command will be documented with the first release.