{"id":25406890,"url":"https://github.com/xynehq/xyne","last_synced_at":"2025-05-15T20:00:40.150Z","repository":{"id":274954085,"uuid":"851256132","full_name":"xynehq/xyne","owner":"xynehq","description":"AI-first Search \u0026 Answer Engine for work. Open-source alternative to Glean.","archived":false,"fork":false,"pushed_at":"2025-05-13T03:38:52.000Z","size":10088,"stargazers_count":521,"open_issues_count":124,"forks_count":39,"subscribers_count":5,"default_branch":"main","last_synced_at":"2025-05-13T04:27:43.389Z","etag":null,"topics":["assistants","bedrock","bun","claude","enterprise-search","gen-ai","generative-ai","information-retrieval","integrations","llm","openai","rag","typescript"],"latest_commit_sha":null,"homepage":"https://xynehq.com","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/xynehq.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-09-02T18:22:03.000Z","updated_at":"2025-05-13T01:40:49.000Z","dependencies_parsed_at":"2025-02-15T10:24:02.320Z","dependency_job_id":"4a0bb78e-5e84-4745-abdd-ad27e2b66d73","html_url":"https://github.com/xynehq/xyne","commit_stats":null,"previous_names":["xynehq/xyne"],"tags_count":7,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xynehq%2Fxyne","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xynehq%2Fxyne/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xynehq%2Fxyne/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xynehq%2Fxyne/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/xynehq","download_url":"https://codeload.github.com/xynehq/xyne/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254414456,"owners_count":22067262,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["assistants","bedrock","bun","claude","enterprise-search","gen-ai","generative-ai","information-retrieval","integrations","llm","openai","rag","typescript"],"created_at":"2025-02-16T06:17:58.583Z","updated_at":"2025-05-15T20:00:39.549Z","avatar_url":"https://github.com/xynehq.png","language":"TypeScript","funding_links":[],"categories":["A01_文本生成_文本对话","NLP","TypeScript"],"sub_categories":["大语言对话模型及数据"],"readme":"![Xyne App](docs/readme/Xyne-Readme.png)\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://github.com/xynehq/xyne/releases/latest\"\u003e\n    \u003cimg src=\"https://img.shields.io/github/release/xynehq/xyne.svg?style=flat-square\" alt=\"Release\"\u003e\n  \u003c/a\u003e\n  \u003ca href=\"https://opensource.org/licenses/Apache-2.0\"\u003e\n    \u003cimg src=\"https://img.shields.io/badge/license-Apache%202.0-blue?style=flat-square\" alt=\"License\"\u003e\n  \u003c/a\u003e\n  \u003ca href=\"https://join.slack.com/t/xynerds/shared_invite/zt-2z2x1c7cw-wMTfymVAhmu2abkVXopVQQ\"\u003e\n    \u003cimg src=\"https://img.shields.io/badge/slack-xyne-brightgreen.svg?logo=slack\u0026style=flat-square\" alt=\"Slack\"\u003e\n  \u003c/a\u003e\n\u003c/p\u003e\n\nXyne is an AI-first Search \u0026 Answer Engine for work. We’re an OSS alternative to Glean, Gemini and MS Copilot.\n\nYour work information has become fragmented — across so many SaaS apps, docs, files, repos, projects, issues, emails, messages, tickets, etc. Xyne connects to your applications (Google Workspace, Atlassian suite, Slack, Github, etc), securely indexes your data, and maps a graph of relationships. So you get a (Google + ChatGPT) like experience to find anything across it all and get up-to-date answers with sources. \n\nFind that file, triage that issue, ask everything about a customer / deal / feature / ticket, discover the right people. I.e. Get the right information quickly to move your work forward.\n\nWe already have intelligent AI models. What's missing to make them work at your work is providing them with your context. In a secure, private and responsible manner. That's what Xyne does.\n\nXyne is the most secure \u0026 future-proof solution to bring AI into your work and your org.\n\nYou can check out our [documentation](https://docs.xynehq.com) to learn more about Xyne.\n\n### Live now on entire Google Workspace\n\nSo you can ask and search anything across not just your drive, PDFs, docs, sheets, slides; but also your calendar, contacts, gmail — including all your attachments! \n\nWhen most Enterprise Search solutions say Google Worksuite integration, they generally mean drive (and conveniently skip contacts, calendar, and gmail).\n\n**NEXT: Slack and Atlassian integrations**\n\n## ✨ Features\n\n- **Self-hosted**: Deploy anywhere - laptop, on-prem, cloud.\n- **Model Agnostic**: Plugs into any LLM of your choice. Also\nvia any cloud provider of your choice. You can even point it\nto a local Deepseek via ollama.\n- **Private \u0026 secure**: No training on your data or prompts. No\ntelemetry.\n- **Permissions-aware**: Live enforcing of your apps' existing\npermissions; no one sees what they don't have access to.\n- **High performance**: Multi Threaded data ingestion\n\n## 📺 Demos\n\nAI Search and Answer engine for Work\n\n[![Alt text](https://img.youtube.com/vi/ZvyRp4o_p-0/0.jpg)](https://www.youtube.com/watch?v=ZvyRp4o_p-0)\n\nDeepSeek over internal data\n\n[![DeepSeek](https://img.youtube.com/vi/rl5J-DeHG04/0.jpg)](https://www.youtube.com/watch?v=rl5J-DeHG04)\n\n## 💻 Setting up Google Workspace integration\nFollow the steps mentioned for the [Service Account integration](https://docs.xynehq.com/authentication/service-accounts) of Google Workspace\n\n## 🚀 Deployment\n\nXyne can be easily run locally or deployed on a virtual machine using a single ```docker-compose``` command.\n\n#### To run Xyne locally:\nTo run xyne locally on your machine, check out the [Quickstart](https://docs.xynehq.com/quickstart) section in our docs.\n\n#### To run Xyne in your Cloud Provider:\nTo run Xyne on AWS EC2, check out the [Cloud Deployment](https://docs.xynehq.com/deployment/cloud/aws/aws-deployment-with-docker) section in our docs.\n\n## 💪🏻 Building the most open, modular, extensible Work AI Platform\n\nWhat augments humans with the right context also augments LLMs (AI Agents / Assistants) with retrieval, memory, and tooling. That's the platform we're building. An Org Operating System (OrgOS) that provides you with the right tooling to build AI-first internal applications and workflows tailored to your \u0026 your team's requirements.\n\n## ✒️  Contribute to Xyne\nIf you want to contribute to Xyne, checkout our [Contribution Guide](https://docs.xynehq.com/contribution/contribute) for more details.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fxynehq%2Fxyne","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fxynehq%2Fxyne","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fxynehq%2Fxyne/lists"}