{"full_name":"langgenius/dify","owner":"langgenius","name":"dify","url":"https://github.com/langgenius/dify","description":"Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.","homepage":"https://dify.ai","topics":[],"language":"Python","license":"NOASSERTION","avatar":"https://avatars.githubusercontent.com/u/127165244?v=4","created_at":"2023-04-12T07:40:24Z","pushed_at":"2026-11-10T09:34:05Z","metrics":{"stars":160819,"forks":25434,"watchers":838,"issues":19760,"prs":20801,"releases":172,"issues_open":null,"issues_closed":null,"prs_open":null,"prs_closed":null,"prs_merged":null,"discussions":4304},"growth":{"day":{"stars":84,"forks":8,"base_day":"2026-11-09"}},"history":{"dates":["2026-11-09","2026-11-10"],"stars":[160735,160819],"forks":[25426,25434],"watchers":[838,838],"issues_open":[null,null],"prs_open":[null,null]},"boards":{"total":58},"same_owner":[],"issues":[{"number":39151,"title":"Parallel fail-fast leaves collateral node failures unattributed and stale","comments":59,"created_at":"2026-07-16T13:10:38Z"},{"number":24645,"title":"[Chore/Refactor] remove redundant-cast","comments":22,"created_at":"2025-08-27T11:41:11Z"},{"number":33092,"title":"[Refactor/Chore] replace json.load by pydantic","comments":20,"created_at":"2026-03-06T09:49:51Z"},{"number":30269,"title":"the roadmap for adapt to modern tech for backend","comments":17,"created_at":"2025-12-28T08:40:14Z"},{"number":32679,"title":"Qwen3.5 reasoning content display","comments":11,"created_at":"2026-02-27T05:36:30Z"},{"number":40874,"title":"v1.16 Agent mode: uploaded files not displayed in UI and file variables invalid in build","comments":12,"created_at":"2026-08-17T08:35:33Z"},{"number":40773,"title":"Model provider plugin migration error","comments":11,"created_at":"2026-08-14T08:42:11Z"},{"number":40680,"title":"Keyword search silently returns no results for parent-child chunks","comments":12,"created_at":"2026-08-13T02:57:50Z"}],"discussions":[{"number":45755,"title":"dify1.18.1, calling the https://api.dify.ai/v1/datasets/{dataset_id}/document/create-by-file API throws an error.","comments":1,"category":"Help","created_at":"2026-11-07T02:20:03Z"},{"number":45739,"title":"Can't dify be deployed through Kubernetes?","comments":1,"category":"General","created_at":"2026-11-06T12:08:09Z"},{"number":45704,"title":"In version 1.18.1, when calling the https://api.dify.ai/v1/workflows/{workflow_id}/run API, I found that the file parameter gives an error. What could be causing this? Because the same parameter format worked fine in version 1.13.3.","comments":2,"category":"Help","created_at":"2026-11-06T07:18:04Z"},{"number":45678,"title":"Show and tell: an HTTP Request node that checks a workflow's draft before anyone sends it","comments":0,"category":"Show and tell","created_at":"2026-11-05T15:10:58Z"},{"number":45592,"title":"In Didy1.13.3, when searching the knowledge base, it seems like metadata filtering isn't working.","comments":1,"category":"Help","created_at":"2026-11-05T02:21:31Z"},{"number":45582,"title":"How do you prove an AI agent's output today is the output that passed your checks?","comments":0,"category":"Show and tell","created_at":"2026-11-04T17:19:43Z"},{"number":45541,"title":"v1.19.0 - Trace Your Workflows, Layer by Layer","comments":0,"category":"Releases","created_at":"2026-11-04T07:36:16Z"},{"number":45305,"title":"Image generation and editing in Dify: should they become first-class model types, or stay Tool plugins?","comments":0,"category":"General","created_at":"2026-11-02T05:39:13Z"}],"latest_release":{"name":"v1.19.0 - Trace Your Workflows, Layer by Layer","tag":"1.19.0","published_at":"2026-11-04T07:36:16Z","body":"**Dify 1.19.0 brings nested workflow tracing to your observability tools.** Follow a workflow into the Workflow Tools it calls, inspect the steps inside each invocation, and keep the LLM context alongside the execution tree.\r\n\r\nThis release also makes Workflow Tools portable with App packages, adds Skill Hub automation, and makes Agent files easier to upload and preview. The bundled plugin daemon is upgraded to **0.6.14-local**.\r\n\r\n[Release highlights](#user-content-release-highlights) · [Upgrade guide](#user-content-upgrade-guide) · [Detailed changes](#user-content-detailed-changes)\r\n\r\n\r\n\r\n## Release highlights\r\n\r\n### Nested workflow tracing\r\n\r\n**See what happens inside a Workflow Tool, all the way down to its LLM calls.**\r\n\r\nBuilding on the Console drilldown introduced in 1.18.0, new external traces now include the persisted execution steps inside called Workflow Tools. Child steps appear beneath the Tool span that invoked them, preserving the hierarchy across nested and repeated calls. (#44673)\r\n\r\n\r\n\r\n*From the outer workflow to a nested LLM call: the execution tree and formatted messages, together in Langfuse. Screenshot from [#44673](https://github.com/langgenius/dify/pull/44673).*\r\n\r\n- **Follow the hierarchy.** Expand called Workflow Tools to see their internal execution steps under the invoking Tool span.\r\n- **Understand each invocation.** Repeated calls and failures retain their place in the trace.\r\n- **Keep the LLM context.** Inspect formatted prompts and responses, token usage, and project, trace, and observation links.\r\n\r\n> [!NOTE]\r\n> **Your saved provider settings continue to work.** The richer hierarchy applies to new executions; historical traces are not backfilled. Independently nested workflow runs remain separate traces linked by parent identifiers. For rolling deployments, follow the [tracing queue transition](#user-content-upgrade-guide) below.\r\n\r\n### Take Workflow Tools with your apps\r\n\r\nExport an app and bring the workflows behind its Workflow Tools with it. `.ifpkg` packages include those workflows at their bound versions, including transitive dependencies. Import reuses tools already in the destination workspace and recreates, publishes, and registers missing ones, remapping their references. (#44592)\r\n\r\nAll Dify app exports now use **package format v2**, including apps without Workflow Tools. Existing v1 imports and explicit YAML export remain supported. See the [upgrade guide](#user-content-upgrade-guide) before importing new packages into older installations. (#45353)\r\n\r\n### Upload together. Preview in place.\r\n\r\nSelect or drop multiple Agent configuration files and upload them in parallel. Successful uploads are bound in one atomic batch, concurrent Composer saves preserve the latest files, and Prompt uploads keep the cursor position while inserting multiple references. (#44947)\r\n\r\nRead PDFs inline, inspect source files with syntax highlighting, and see when text has been truncated. Previews cover configured files, Skill members, and working-directory files, with a download action when a PDF cannot be loaded. (#45032, #45045)\r\n\r\n\r\nSee Agent file uploads and previews
\r\n\r\n\r\n\r\n*Read a PDF directly in the Agent file preview and switch between project files.*\r\n\r\n\r\n\r\n*Select several files and upload them together.*\r\n\r\n![Syntax-hi"},"readme":true,"readme_zh":false,"news":[{"title":"Dify, a visual workflow to build/test LLM applications","url":"https://github.com/langgenius/dify","hn":"https://news.ycombinator.com/item?id=40121318","source":"Hacker News","lang":"en","date":"2024-04-22","points":185,"comments":38},{"title":"Dify 1.0","url":"https://github.com/langgenius/dify/releases/tag/1.0.0","hn":"https://news.ycombinator.com/item?id=43218661","source":"Hacker News","lang":"en","date":"2025-03-01","points":2,"comments":0},{"title":"Dify, an end-to-end, visualized workflow to build/test LLM applications","url":"https://github.com/langgenius/dify","hn":"https://news.ycombinator.com/item?id=40110547","source":"Hacker News","lang":"en","date":"2024-04-22","points":2,"comments":0},{"title":"Dify: Production-ready platform for agentic workflow development","url":"https://github.com/langgenius/dify","hn":"https://news.ycombinator.com/item?id=47313644","source":"Hacker News","lang":"en","date":"2026-03-09","points":1,"comments":0},{"title":"Dify's Open Source License","url":"https://github.com/langgenius/dify/blob/90bc51ed2e6672c4a4d6c0199d218aa87037810a/LICENSE","hn":"https://news.ycombinator.com/item?id=43482644","source":"Hacker News","lang":"en","date":"2025-03-26","points":1,"comments":1},{"title":"Dify 1.0: Plugins, Marketplace, a New Agent Node, Optimized RAG and Memory","url":"https://github.com/langgenius/dify/releases/tag/1.0.0","hn":"https://news.ycombinator.com/item?id=43240627","source":"Hacker News","lang":"en","date":"2025-03-03","points":1,"comments":0},{"title":"How Dify Experience is Viewed in Projects | The Work of Building Generative AI Apps with Low-Code and the Required Experience","url":"https://note.com/fdeai/n/n66d0947ed315?hl=en","source":"note","lang":"en","date":"2026-11-01"},{"title":"n8n, Dify, and Ollama might be the best self-hosted AI automation stack right now","url":"https://www.xda-developers.com/n8n-dify-ollama-best-self-hosted-ai-automation-stack/","source":"XDA Developers","lang":"en","date":"2026-04-13"},{"title":"Dify Secures $30M Series Pre-A to Power Enterprise-Grade Agentic Workflows","url":"https://theaiinsider.tech/2026/03/10/dify-secures-30m-series-pre-a-to-power-enterprise-grade-agentic-workflows/","source":"theaiinsider","lang":"en","date":"2026-03-10"},{"title":"Dify Raises $30 million Series Pre-A to Power Enterprise-Grade Agentic Workflows","url":"https://venturebeat.com/business/dify-raises-30-million-series-pre-a-to-power-enterprise-grade-agentic-workflows","source":"VentureBeat","lang":"en","date":"2026-03-09"},{"title":"Dify 深耕香港市場","url":"https://biz.hkepc.com/24115/Dify_%E6%B7%B1%E8%80%95%E9%A6%99%E6%B8%AF%E5%B8%82%E5%A0%B4_%E7%9E%84%E6%BA%96%E4%BC%81%E6%A5%AD%E7%B4%9A_Agent_%E6%87%89%E7%94%A8%E7%99%BC%E5%B1%95","source":"biz.hkepc.com","lang":"zh","date":"2025-09-04"},{"title":"The story from finding Dify on Github to receiving a promotion code","url":"https://note.com/conaxam/n/n15fe97ab6a9c?hl=en","source":"note","lang":"en","date":"2025-04-03"},{"title":"实测筛选：三个可直接落地的GitHub开源项目（Dify/DataV/PlatformIO）","url":"https://news.google.com/rss/articles/CBMiX0FVX3lxTFBQQ2RWcFVwa0xRTlpZenlMM1RINHg3UGxIaTdnNmtDMTZkNnBoTTBnMEY4UDBNT2dQY0xBaW1ENU5CTm5uSmhjSVE5WUJ1S3VYcWRSaklIOFdvM0VscGEw?oc=5","source":"CSDN","lang":"zh","date":"2026-11-08"},{"title":"Dify vs n8n: Which Platform to Choose for AI Apps (and How to Use It with OpenRouter)","url":"https://news.google.com/rss/articles/CBMie0FVX3lxTE53SElUQUUyb1ZpSGRTQUdLRkxpRDVub0EwOW00QzZuNW5PTWhGbTk5aUdfRURtYkNRcFpEaUJFOTZxcVBtS0I0N1FUZ0IwTkVWSFVMUUM1M29XZ1gxU3Axc2V3SmJtSEtCLTVuV0VWY2dScDVjYl9nOWllRQ?oc=5","source":"Pasquale Pillitteri","lang":"en","date":"2026-06-25"},{"title":"Researchers Detail DifyTap Flaws in Dify That Could Expose AI Chats Across Tenants","url":"https://news.google.com/rss/articles/CBMif0FVX3lxTFA1dmpzcFV6Wl94djJ2Nmg0dmgzc250VlhzRVpqOG9EdTdLa2I2eEJCVWQxaTlpSEhiUjRtb0VpLVhkUG5kOFhURWZQOXlpQ3RHMXpKYkdXVGFkdDBIVkExcnVvTTdVbGdka2FXRWZLcVRONFB2NGNiT2ZUc1BsN1E?oc=5","source":"The Hacker News","lang":"en","date":"2026-06-22"},{"title":"Dify, popular AI platform with 10 million installs, exposed users to one-click account takeover","url":"https://news.google.com/rss/articles/CBMie0FVX3lxTE5kR2o5ZmkwOS00Qi1FZEdIdlFtRjdkWHdzcHFBQXdVZEttT3N2dlFCT3FZTFJBTmxySk9RcUxRUGlLRlNMSExXWnd1eDNHVFZWaDdRSmJHamJlaXdfSjBlOWF1NXpRS0hDaDhOek5oMzIzS0J1anpsR2xtaw?oc=5","source":"Cybernews","lang":"en","date":"2026-05-21"},{"title":"Dify raises $30M to scale AI workflows","url":"https://news.google.com/rss/articles/CBMisgFBVV95cUxNQ2tGM1JSMmpSX3VraTFPekZ6eExfU0lhRWJwaHpmWmEtWndqalpOTko5NzBIOFBsTmI2Y05lUGdZc3NKY21jOTlJN1FIOF9RZE5vTjBidWtpV2M2ZjN0UjQtOUpMcGJxZTNrUk1SQ1hibjNnMFJZd1EzUjNON0hLb1UxcjZUVEdrR2dNT2tfUTlZRGVvZm5MZUx6TUE4VVRkMk9nTDdaWjBlYWRXa09icnpB?oc=5","source":"VentureBeat","lang":"en","date":"2026-03-09"},{"title":"Dify 自建部署完全指南：从上手到放弃到真香","url":"https://news.google.com/rss/articles/CBMiXkFVX3lxTE1vbXMxaDNIR1p1bUNLUkFzT2RjWkk0UXlONXRNenB6TTJKY0pZa1dYRUZoUTdzR292UW5FaElDVU9XOXEyMzFIMlpEX2oxMzBVSDZiZFhWdXRpUzhtSkE?oc=5","source":"53AI","lang":"zh","date":"2025-11-20"},{"title":"张路宇：我们专门招大厂不要的人，Dify赢在组织方式上","url":"https://news.google.com/rss/articles/CBMia0FVX3lxTE9pdi1rLUpBZWlnRWtvM1N0OGlRYzNKZ2xzNE8tNWt0SlhUWUFBVzZWLTZLZzFfajVrZDdkc1dVTnE1S2Iwd2RPQnB4T2hHdVY3ajVjdE5nQmEtT0FFSzY3RlFOTHFralViWGVJ?oc=5","source":"财富号","lang":"zh","date":"2025-11-07"},{"title":"Dify Quickstart Guide: Build Your First AI Workflow","url":"https://news.google.com/rss/articles/CBMiXEFVX3lxTE5ncmlHbUFiWkZ2S2JISF9sYjBScjRmT3BUcnQyaW94OVpFT0V4NnFrdVpQWU5jMjRBcngzVkg5YWlWUVFPeVI3LUFpQnhkbWVYcXkyRFc5ckU1ODFl?oc=5","source":"YouTube","lang":"en","date":"2025-09-18"},{"title":"上市企业向 Dify 核心贡献者发送辱骂邮件","url":"https://news.google.com/rss/articles/CBMiS0FVX3lxTE95OHM1VlQ5Y2t3R1dBejFrQjJBajY5UzNnVWJVb2g5alN0NDRQNjNZNjNCV1VBVzdsWlkyWFhLdVZrZ2VhZHdOandBRQ?oc=5","source":"OSCHINA","lang":"zh","date":"2025-03-28"},{"title":"打开日本市场背后，Dify 是怎么做 AI 全球化的？","url":"https://news.google.com/rss/articles/CBMif0FVX3lxTFBDeXVHYTFGcGdJQW9Gb240M2xvMjdjcXdSWXJnd2VkdTB3SUxnQnV0UlpNYjdfdllSdEJjdVRySllKZ2VoTS1jTjYzeXRnWlp5bWhnUkFOQmtISjlnVWZMektHYzR3TDd1OVNqejh3RkwtYlo1MW10SndPQVB5N0U?oc=5","source":"新浪财经","lang":"zh","date":"2024-12-18"}],"stale":false,"snapshot_at":"2026-11-10T09:35:25Z","history_start":"2026-09-06"}