Back to directory

MCP Lens for DeepSeek Harness

A DeepSeek Harness plugin that replaces loading every connected MCP server's full tool schema with two model-facing tools, one that searches across servers for relevant tools and one that calls a specific tool by name.

Visit website (opens in a new tab)deepseek-harness-mcp-lens.charmingkla.chatgpt.site
MCP Lens for DeepSeek Harness

MCP Lens gives DeepSeek Harness two tools: one finds relevant tools and returns their exact input schemas, and the other invokes a specific server and tool and returns its result, including structured data that stays available for later calls in the same conversation.

In a 1,000-tool test fixture, the two standing tool definitions occupy about 1.1 kilobytes of JSON regardless of catalog size, compared with roughly 648 kilobytes for a direct client that loads every schema up front. Remote schemas only enter the conversation once a search returns them, connections open on demand, and repeated searches reuse the catalog index.

The plugin targets catalogs of dozens to thousands of tools spread across many MCP servers; for a handful of tools used on nearly every request, the project recommends the official direct MCP client instead. It requires Node.js 22.19 or newer and a matching DeepSeek Harness version.

More in MCP servers

See all 734
  • Light Cloud

    light-cloud.com

    A hosting platform that deploys a static site, a container, or a managed Postgres database straight from a GitHub, GitLab, or Bitbucket push, with a preview URL for every branch and an MCP server for deploying from Claude Code.

  • ForensicDbg

    forensicdbg.com

    A post-mortem debugger for native Windows x86 and x64 crashes, with an MCP interface so AI tools can consume crash-dump analysis directly.

  • Karada.ai

    karada.ai

    A CI/CD platform that auto-compiles existing APIs into Go MCP servers, attaches composable middleware, and exposes them through one gateway usable from Cursor, Claude Code, and Windsurf.