关于这个工具
Bag of words 是一款开源 AI 分析师,旨在将任意 LLM 连接到任意数据源,并实现集中式上下文管理、信任、可观测性与控制。 其核心能力围绕“知识挂具”(Knowledge Harness)构建,为智能体提供可用的上下文:包括限定范围的连接、表与语义对象、文件、工具、指令、身份、权限以及历史证据。智能体可跨数据库、服务、API、表、语义对象、文档和文件开展工作。 该产品支持从 dbt、Tableau、代码、AGENTS.md 等来源摄取数据,并让 AI 持续维护和监控上下文。AI 会根据反馈和用户请求自动建议对上下文的改进与更新。 团队可以创建自定义指令和 AI 规则,并配以审批工作流。评估与反馈形成自我改进闭环,使团队能够审查并推广更好的指令,而不是让智能体在无人监督的情况下自行更改规则。 Bag of words 适用于业务分析(收入与财务分析、运营审查、自助探索)以及技术调查(跨可观测性系统、服务和数据库的生产根因分析)。它提供高层性能指标,包括上下文覆盖率和 AI 准确率。 该产品面向业务分析师、数据团队和工程师,学习曲线为中等,并以开源形式发布。
功能与使用场景
Agents can work across databases, services, APIs, tables, semantic objects, documents, and filesAI will auto suggest improvements and updates to your context based on feedback and user requestsBusiness analysis: revenue and financial analysis, operating reviews, and self-serve exploration.Collect and track user feedback to improve AI accuracyCreate custom instructions and AI rules with approval workflowsEvals and feedback close the self-improvement loop, so teams can review and promote better instructions rather than giving an agent unchecked authority to change its own rulesFrom question to defensible result Bag of words runs a governed agentic loopGet high level performance metrics including context coverage, AI accuracyIngest from dbt, Tableau, code, AGENTS.md, and have AI continiously maintain and monitor your contextIntegrate to your existing tools like dbt, lookml, AGENTS.md, and moreIts Knowledge Harness is the context that makes the loop useful: scoped connections, tables and semantic objects, files, tools, instructions, identity, permissions, and prior evidenceManage agents Shape agents with data, tools, files, instructions, and evals
使用分析
信息来源
每条档案都保留来源与最近验证时间。相似工具
Docs by LangChain
docs.langchain.com
90
Deploy and manage agents with durable execution, real-time streaming, and horizontal scaling.
免费增值 · US$1+/月
Cursor
cursor.com
90
Cursor 是一个 AI 驱动的编程平台,帮助开发者通过自主智能体构建软件。它支持桌面、网页和移动端,可在终端、Slack 和 GitHub 中运行。
付费 · US$8+/月
