well
这是什么
智能体的上下文窗口很小,而它的记忆只是猜测。the well 就是同时解决这两个问题的器官:往里面倒入多少内容都行(它存在磁盘上,不在提示词里),把最接近的有依据段落取回来——而最关键的部分是——当字段实际上并不包含你所问的内容时,它会诚实地返回 ABSENT。
只有三个动词,而且只有三个:
feed(text | file) pour content into a universe (unlimited external context)
recall(query) draw the nearest grounded passages back out, with scores
ask(question) PRESENT with grounded evidence, or ABSENT — no confabulationthe well 不生成语言。它负责存储、检索和弃权。生成属于饮用它的智能体;the well 的职责是确保智能体只饮用真正存在的东西。
诚实之处
弃权正是记忆系统悄悄撒谎的地方——它们总是返回某个东西,而某个东西看起来总像答案。the well 通过构造实现弃权,这条规则是对 ΑΔΩ 研究发现的忠实解读,即缺失是信号的缺乏,而不是低分:
只有当字段中存在一个极点时,查询才是有依据的——最佳匹配必须与查询共享一个真实的内容词并且高于字段自身的背景。仅凭字符纹理永远不足以构成依据。当最佳匹配无法从噪声中分离时,字段就是一片平坦的高原,而平坦的高原意味着 ABSENT。
这一点也诚实地承认了自身的上限:v0.1 是一个轻量、无依赖的词汇字段——真实的召回、真实的弃权,但不是语义理解。它是研究推向 1.00 召回率和校准弃权的基础,也是衡量接近程度的测试平台——而不是第一天就宣称这些数字的黑箱。
使用它
well feed -f ./notes.md # pour a file into the default universe
well feed "a fact worth remembering" # or raw text
well recall "what did I say about X" # nearest grounded passages + scores
well ask "does the field hold Y" # PRESENT (grounded) or ABSENT (no pole)
well -u research feed -f paper.txt # named universes run side by side
well stats # what a universe holds每个宇宙都是 $WELL_HOME(默认 ~/.well)下的一个紧凑文件。想运行多少就运行多少——这是启动多个字段系统的开端。
作为 MCP 工具
the well 通过 stdio 使用模型上下文协议(零依赖),因此 AI 可以在思考过程中从中取用。用 Claude Code 注册它:
claude mcp add well -- node /path/to/well/mcp/server.js它暴露了 well_feed、well_recall、well_ask 和 well_stats。此后,智能体可以喂入一个宇宙,然后向它提问,获得有依据的证据或诚实的 ABSENT——而无需将语料库加载到其上下文窗口中。这就是重点:智能体在后台运行一个宇宙,而人们仍然正常地与它交谈。
与 kith 同行
the well 是记忆器官;kith 是智能体相遇的桌子。它们完美契合。一个由 well 支撑的席位(adapters/kith-well.js)只在字段为某个主张提供依据时才 COMMIT,并在字段干涸时投出闸门的 NOTHING——一种真实的、经过测量的弃权。kith 的"我不知道"不再是猜测,而变成了"不在字段中"。
node test/well.test.js # every primitive, offlineThis server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Shared, governed long-term memory for AI agents across tools and sessions via MCP and REST.
Shared long-term memory vault for AI agents with 20 MCP tools.
Private-by-default, local-first memory/context/task orchestrator for MCP apps and agents.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/AlphaDataOmega/well'
If you have feedback or need assistance with the MCP directory API, please join our Discord server