labmimors/dsh-mcp-lens

README 摘要

MCP Lens for DeepSeek Harness English 简体中文 Shrink large MCP catalogs to a two-tool model surface. MCP Lens lets DeepSeek Harness search and call 1,000 remote tools through two stable model-facing interfaces. Instead of sending every tool schema on every turn, it reveals exact schemas only for a small ranked set when a tool is actually needed. Why users install it: - Spend less on input-heavy turns: in the dated three-task pilot, estimated DeepSeek V4 Flash cost fell from $0.0307204 to $0.0034707 . - Keep more room for the real task: the same pilot reduced request/header.tools JSON from 674,249 B to 27,401 B . - Retrieve more relevant covered calls: on a frozen MCP-Atlas-derived convenience holdout, Recall@5 rose from 0.062610 to 0.246656 across 304 untouched prompts. This is lexical retrieval evidence, not an official MCP-Atlas or end-to-end score. - Avoid rebuilding the same search index on every query: rc.9 reuses the tokenized index for each Lens-owned frozen catalog and policy generation, then invalidates it when the catalog changes. - Narrow the tool-choice surface: search reveals only a small ranked set of exact schemas, and the final server/tool is still gated by allowTools …

分类