Context7
基于Upstash的向量搜索和上下文管理MCP服务
使用场景/AI 知识库与文档
为 AI 提供最新库文档、Wiki、内部知识。减少幻觉,适合写代码、学新框架。
共匹配 1,994 个资源 · 第 25 / 42 页
知识库类 MCP 解决大模型最大的短板——知识过期与幻觉:Context7 按库与版本实时拉取官方文档和代码示例,DeepWiki 把任意 GitHub 仓库编译成可问答的 Wiki,自建方向则常用向量库 / 知识图谱 MCP 接入团队内部资料。
写代码时的典型链路:在 Cursor 里问「Next.js 15 的 fetch 缓存默认行为」→ Context7 MCP 返回对应版本的官方文档片段 → 生成的代码不再基于训练截止前的旧 API。升级大版本、试用新框架时收益最明显。
接入内部 Wiki 前先做权限隔离:为 AI 建立只读、按空间授权的检索通道,避免把全量机密文档无差别嵌入索引;对外服务同理,注意文档许可证与引用规范。
v0.1.2
Bonanza MCP — 'Bonapitit' API (api.bonanza.com/docs), search + item detail
v0.4.0
Send tasks and notifications to people's phones. Answers build a queryable knowledge base.
v1.3.0
Issue tracker and wiki for teams and their agents: find, file and update issues, write pages.
v0.1.0
Document conversion and OCR for AI agents: PDF, Office docs, images to text.
v2.0.0
Design system contracts, docs search, and usage validation for @digitaltableteur components.
v0.3.7
Find APIs and MCP servers for AI agents: auth, free tier, MCP endpoint, docs, last-checked date.
v1.0.0
Tasks, sprints, releases and docs for engineering teams with AI coding agents.
v1.0.0
Evidence-first entity and knowledge graph MCP server for AI agents to verify claims.
v0.13.4
Memory, docs, and repo context layer for engineering agents.
v1.0.0
Search and edit your company's FakeParrot wiki and run its workflows, as the signed-in employee.
v0.5.3
Graph memory that dreams: Neo4j knowledge-graph memory for AI agents with hybrid search
v0.4.0
One place for every AI agent's pages and docs: versioned links to share, search and update.
v1.3.0
Crypto perp funding, spreads and listings, plus your own Arbitron cards, trades, PnL and docs.
v1.0.0
Guru knowledge base search and collections.
v1.0.1
stdio MCP server wrapping the Redmine REST API: issues, projects, users, time, wiki, versions.
v2.3.0
开源的双时态 LLM 智能体记忆层——带冲突消解的知识图谱。
v1.1.0
Find and manage Clearly workspace docs, canvases, sheets, decks, projects, boards, and tickets.
v1.0.0
Official facts about Lovora (18+ AI companions): companions, prices, content rules, docs.
v1.0.0
Placement across agent channels: MCP, Cursor, Claude, ChatGPT, npm, docs, Smithery, Glama.
v1.0.2
Names every image, font, badge and embed in your docs that sends an EU reader's IP outside the EEA
v0.22.3
Print your business letterhead on Word docs, proposals and contracts from chat.
v0.6.2
Your reviewed knowledge, across AI conversations.
v0.1.0
Find tools and hosted agents by task, with source documentation and connection requirements.
v0.29.2
面向 agent 的 MIND 知识图谱记忆,44 个工具:CRM、生活、MoneyMIND、预算、表格、邮件与签名。
v2.1.2
图谱驱动的代码智能,为 AI 助手提供语义搜索与知识图谱。
v1.0.1
把一个产品 URL 变成带旁白的电影感演示视频、发布视频或演示稿。
v1.0.2
Critical dates and deadlines in leases, loans, policies, HOA docs and contracts, with the clause.
v4.12.0
跨域知识网络,AI 智能体通过 MCP 搜索、共享并打磨解决方案。
v1.3.0
Local-first MCP server giving AI agents contextual knowledge about the people in your life
v0.5.0
Cognitive memory for coding agents: hybrid recall, remember/forget, knowledge graph, nightly dreams.
v1.0.0
Use this MCP server to textfiles.com, the BBS era archive of zines, phreaking docs.
v1.0.0
Keyless MCP server for Hex: Elixir and Erlang package versions, docs, downloads, releases.
v1.0.0
Search the MITRE ATT&CK enterprise knowledge base for adversary techniques and tactics.
v1.0.0
Search the MDN web docs for HTML, CSS, and JavaScript. No key required.
v1.0.1
Use this MCP server to the Dwarf Fortress wiki, !!FUN!!
v1.0.0
Cited, versioned knowledge for agents: retrieve sourced passages and propose owner-approved fixes.
v1.0.0
Dentistry knowledge base: implantology, periodontology, oral surgery. Verbatim text, with sources.
v11.9.8
通过 MCP 读取 MindForge 引擎:知识图谱、项目健康度与审计日志。
v1.0.1
Search, read, and safely update Markdown notes in your connected Phasoric knowledge vaults.
v1.7.2
Indexes your project's markdown docs and serves them to any agent via local hybrid search.
v1.0.2
Ask your team's code, Slack and docs from any MCP client. Every answer cited to source. Early access
v0.8.5
零知识密钥管理:本地加密面 MCP 服务,负责解密并注入密钥。
v1.0.0
Personal knowledge MCP: capture bookmarks, notes & todos by chat; archive pages; search memory.
v1.1.0
Build, test, deploy and run AI phone agents: agents, numbers, calls, tests, knowledge, campaigns.
v1.0.0
Create courses, search approved knowledge, preview learning content, and manage learning sites.
v1.0.0
Persistent knowledge graph for AI agents. Remember, recall, and forget facts.
v0.1.0
Run WhatsApp AI agents: manage agents, conversations, CRM contacts, orders and RAG docs.
v1.0.2
Low-cost live web search for current facts, news, research, docs, and web grounding via x402.
Context7 以库名+版本精确取文档切片并结构化给模型,省去抓取清洗、上下文更聚焦;直接搜索适合无固定库的泛调研。
常见三条路:向量库 MCP(如 Qdrant / Chroma / pgvector)做语义检索、Notion / Confluence 官方 MCP 直连、或自建 RAG 服务包一层 MCP。按数据敏感级别选。
会增加一次工具调用的延迟(通常几百毫秒),但换来更少返工。控制返回片段大小、只挂载需要的知识库即可。