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 26 / 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。
v0.3.0
MCP server for LINDAS — the Swiss administration's linked-data SPARQL knowledge graph (cube.link)
v1.0.0
Governed map of company knowledge for humans + agents.
v1.2.1
面向 AI 智能体的项目管理:任务、文档、决策与工时共享同一团队上下文。
v1.0.1789797106
MCP server for learnhouse documentation, generated by doc2mcp.
v0.1.2
Zero-model recall over your docs: the section that answers a question, in about a millisecond.
v1.5.1
为 AI 智能体提供视频转码、规格统一与拼接、文档转换和音频转写,提供免费 1 美元试用。
v0.4.0
面向 AI 智能体的 Cities: Skylines II wiki、百科与模组工作流工具。
v3.7.9
Search @imqueue docs and scaffold typed services & clients from your AI coding agent.
v0.11.0
Local codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
v0.3.0
Self-hosted MCP memory server: a typed knowledge graph on Postgres 16 and pgvector.
v0.8.1
Search self-contained .pikelet knowledge artifacts over MCP — no vector DB, no embedding API.
v0.1.0
Evidence-backed Android repository knowledge for humans and AI agents
v1.0.0
Agile project management over MCP: boards, epics, roadmap, retros, whiteboards, wiki, metrics
v1.0.0
Shared memory for coding agents and their teams: docs, epics, tasks and decisions over MCP.
v0.1.0
Task management for people and AI agents, with scoped OAuth access to issues, projects, and docs.
v1.0.0
Read-only MCP for AI usage profiles, leaderboards, stats, and docs; no writes or private data.
v0.11.0
面向 AI 智能体的代码库知识图谱 — 支持 162 种语言、亚毫秒查询、少用 99% 的 token。
v0.29.5
为 AI 智能体提供语义记忆:混合检索、知识图谱与记忆整合。
v0.4.0
Cross-session memory and hybrid RAG over your notes and docs. Local-first, with cited answers.
v1.42.4
MCP 指令索引服务器,支持搜索、增删改查、校验与跨仓库知识沉淀。
v1.0.0
Official ImageKit developer server: search the docs and build transformation URLs. No sign-in.
v0.2.0
Free CRE loan sizing, DSCR stress tests, payments, closing costs and cited lending knowledge.
v1.26.0
Reliability intelligence: production incidents, risk register, controls catalog, knowledge graph.
v1.7.2
面向 AI 智能体的零知识 MCP 密钥保险库:密钥在运行时注入,模型始终看不到明文。
v0.24.0
AI 智能体的认知记忆,支持语义回忆、知识图谱与矛盾检测。
v0.2.0
Securely query and analyze business data, dashboards, projections, alerts, and knowledge.
v0.1.1
PIL: local-first knowledge base for your Instagram saved posts, with a read-only MCP server.
v1.0.0
Find Power Apps PCF controls, read their docs, and validate pcfhub.json manifests.
v0.1.0-alpha.4
Offline self-checking to reduce AI-style wording in source code and documentation.
v1.1.6
The Matra rich text editor's documentation as an MCP server: every extension, command and API.
v0.1.3
Search hub knowledge and memories, read saved context, and create or update hub content.
v0.1.3
Decide whether an agent should reuse cached URL knowledge or fetch the resource again.
v3.0.0
A living knowledge graph to read, think against, and leave a deposit in that outlives you.
v1.0.0
Public Fabricate pricing and documentation tools.
v0.37.0
为 AI 工作流提供持久化知识图谱与 MCP 服务器,并附带由 git 跟踪的项目上下文。
v1.0.0
Files what you learn into a personal wiki and quizzes you before you forget it.
v0.4.3
Verified MCP troubleshooting, shared agent knowledge, persistent files, and durable identities.
v1.0.0
Read-only docs for the ADITUS event-technology APIs (ticketing, access, BI) and Shop Micro Frontend.
v0.1.4
Team conventions, anti-patterns, and domain knowledge from .codehabits/ for AI coding agents.
v0.1.0
Shared knowledge cache for AI coding agents — reuse an answer once it exists.
v1.1.0
搭建并运行产品内 AI 助手:小部件、知识库、MCP 连接与用量统计。
v1.1.3
经过验证、标注日期的事实知识库,收录你的 LLM 容易答错的内容。
v0.17.5
零配置、可检索文档的 MCP 服务器,支持 SQLite 或 PostgreSQL。
v0.2.2
提供架构洞察、安全分析与持续更新的文档,含 14 个 Mermaid 工具。
v0.10.5
Local-first memory for AI coding agents: shell history with exit codes, git diffs, docs, MCP.
v1.1.0
Aztra workspace: projects, issues, sprints, goals, wiki and time tracking.
v1.0.1
面向代理的 Telegram 原生 CRM:联系人、商机、销售管道、营销活动、文档与报表。
v0.4.0
Queryable UniFi API docs: endpoint search, schemas, code examples. No controller access.
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。