DeepSeek Capability Hub
Capability Hub 是一个为 DeepSeek Harness 及其他 MCP 客户端设计的惰性 MCP 和技能代理。宿主只看到一个紧凑的工具 capability_hub,而不是为每个已配置服务器提供的每一个工具 schema 承担上下文开销。
智能体可以搜索轻量级目录、检查权限、唤醒一个受信任的服务器、通过 hub 调用它,然后再将其关闭。技能正文只在选中之后才会加载。
为什么存在
大型静态 MCP 配置会浪费上下文,也会让工具选择变得更加杂乱。Capability Hub 保持模型可见的界面稳定:
search → inspect → enable → tools → call → disable一个固定 schema 始终保留在 Harness 提示词中。
子工具的 schema 在请求之前不会进入模型上下文。
tools默认返回名称和描述;完整 schema 为可选。MCP 进程按需启动,并且只在 hub 进程生命周期内存在。
技能按元数据发现,并一次加载一个。
第三方新增内容进入人工审批队列;模型不能自行批准可执行代码。
Related MCP server: capability-router
快速开始
要求:Node.js 22.19+ 和 pnpm。
git clone https://github.com/NeoXider/deepseek-capability-hub.git
cd deepseek-capability-hub
pnpm install --frozen-lockfile
pnpm test
pnpm client -- --json '{"action":"search","query":"demo"}'默认目录中只包含一个内置的 echo MCP 和一个示例 ML 技能。测试不会下载或执行第三方软件包。
DeepSeek Harness 配置
构建 hub,然后将 examples/dsh/cordis.patch.yml 合并到当前使用的 Harness Web 配置文件中,并调整代码仓库的绝对路径。然后重启 Harness。
pnpm buildHarness 将暴露一个面向模型的工具:
mcp__capability_hub__capability_hub从以下命令开始:
{"action":"search","query":"web research"}面向模型的约定
公共 schema 特意保持扁平,以便 LM Studio 等受约束解码引擎可以可靠地编译它。任意子参数以 JSON 字符串传递。
发现并调用一个工具:
{"action":"tools","name":"web-search-neo"}{
"action": "call",
"name": "web-search-neo",
"tool": "web_info",
"argumentsJson": "{\"topic\":\"search_status\"}"
}可用操作:
操作 | 用途 |
| 搜索紧凑的能力元数据 |
| 查看一个能力、权限、配置和环境状态 |
| 通过 |
| 启动或停止一个受信任的 MCP |
| 列出子工具;schema 保持可选 |
| 通过 |
| 加载一个已批准的本地技能正文 |
| 将来自 |
| 列出待处理的提案 |
| 重新加载已批准的目录状态 |
真实集成示例
已包含可供审核的提案示例:
Web Search Tool — 动态网页搜索和浏览器工具。
Unity CLI MCP — 官方 Unity CLI 传输。其工具列表仅在 Unity Editor with Unity Pipeline 连接时才会填充。
这些文件只是示例,不会作为默认可信内容被静默应用。审批之前请认真检查路径、版本和权限。
人工门控安装
模型生成的提案存储在 data/state/pending 下,且不能执行。需要由人工单独运行命令批准:
node dist/src/admin.js approve <proposal-id> --state .\data\state --yes然后调用 catalog.reload 和 enable。优先使用固定版本或不可变的 Git 修订版,避免在已批准的条目中使用浮动的 latest 安装器。
目录、密钥与技能
MCP 传输:
stdio和streamable-http。模板:
${catalogDir}、${packageDir},以及明确列出的${config:key}值。密钥:仅允许环境变量引用。类似密钥的模型配置键将被拒绝。
技能:仅限已批准的本地 Markdown 文件,按需加载,最大 256 KiB。
状态:运行时配置和提案会被 Git 忽略。
信任边界请参阅 SECURITY.md。
当前范围
工具调用会被代理;子 MCP 资源和内置提示尚未桥接。
重连是显式的:先
disable,再enable。远程技能下载、签名验证和沙箱安装器是未来的工作。
拥有不受限制的主机外壳访问权限的模型可以绕过插件局部策略;请使用 Harness 权限作为外部边界。
配套项目
想要一个紧凑的动画桌面视图,展示智能体、上下文、模型、推理和聊天?请查看 DeepSeek Harness Widget。
贡献
欢迎提交 issue 和聚焦的 pull request。新的集成应包含一个固定版本的示例、一组严格的权限描述,以及一个端到端测试。
MIT © NeoXider
Available Tools
1 toolcapability_hubLazy MCP and skill capability hubA
Search and inspect a compact catalog, enable a trusted MCP only when needed, list its tools, proxy a tool call, or load one skill body. Shortest path: action "call" with name, tool and arguments starts the server by itself; use action "tools" (optionally with a query) first only when you do not know the tool name, and "enable" only to start one deliberately. Long results arrive in pages: continue with action "page" and the cursor printed on the last line. Avoid includeSchema unless you need the argument shape. Third-party proposals require human approval outside this MCP. Available capabilities — demo-echo (mcp): Bundled local MCP used to verify dynamic discovery, enable, tool listing, calls, and disable. Skills — before starting a task, load each skill below that fits it with action "skill.load" and its name, then follow it: ml-experiment-review: Review an ML experiment as a falsifiable, reproducible evidence package.
Inputs: action (required), kind, name, tool, +5 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 1 tool
There is only one tool, so there is no cross-tool ambiguity, but it multiplexes several distinct actions (catalog search, enable, tools, call, page, skill.load) behind one name. The description gives a shortest-path heuristic, which mostly mitigates confusion between internal actions.
The single tool name `capability_hub` follows a clear snake_case convention. Internal action names are mostly plain and readable, with only the dotted `skill.load` as a minor stylistic deviation.
A single meta-tool is plausible for a capability hub intended to minimize surface area, but packing catalog inspection, enabling, tool listing, proxying, paging, and skill loading into one tool makes the surface feel thin and overloaded. It is borderline rather than clearly wrong.
The hub covers discovery, enablement, tool listing, calls, pagination, and skill loading, which is a solid lifecycle. However, disabling is mentioned in the bundled demo context but not exposed as an action, and catalog search/inspection actions are implied rather than explicitly enumerated.
Maintenance
Related MCP Connectors
Search, vet & assemble MCP servers from your agent: verified tools, risk labels, and trust scores.
Search and inspect source-pinned agent skills, plugins and tools through a free public MCP.
Agent-native catalogue of Baseframe Labs dev tools and MCP servers.
Find, vet, and run MCP tools through a secure audited gateway with prompt-injection risk scoring
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables agents to securely discover and invoke a centrally governed catalog of tools from distributed internal and external providers, with policy enforcement, quotas, inspection, and audit controls.-
- AlicenseNot gradedqualityCmaintenanceEnables agents to search a local catalog, load skills, inspect MCP tool schemas, and invoke selected tools through a single unified MCP tool, reducing per-request overhead.1MIT
- FlicenseNot gradedqualityBmaintenanceGives AI agents a single MCP interface to discover, inspect, scope, and execute external capabilities—downstream MCP servers, HTTP APIs, and local tools—resolved on demand from a local index instead of preloaded catalogs. Enforces agent scopes, allow/ask/deny policies, health gates, and schema snapshots at execution time, with local execution traces.-
- AlicenseNot gradedqualityCmaintenanceEnables host agents to resolve tasks against a local library of reviewed Skills and Experts, then retrieve selected capabilities on demand through MCP tools without loading the full catalog into context.1MIT