See it in action
An overview from the Titanic demo
Author replay output from an open-ended question about the supplied Titanic CSV: an overall-survival pie chart and a gender comparison. The matching titanic.csv attachment is required.
Example input
What interesting findings can be made from this dataset?Example output
The matching dataset is bundled at user-data/uploads/titanic.csv beside the linked source thread. These are author replay artifacts, with no rerun or independent statistical verification by this site.License
Exact complete human question from the pinned public thread, paired with that same thread’s committed output image and CSV attachment. The trace uses Python/matplotlib/seaborn; it is not an image-model generation. No model reasoning text is republished. The trace does not establish invocation of a named data-analysis Skill, so this example is bound only to the accepted DeerFlow Product. Original author PNG normalized to lossless WebP at its native dimensions. No cropping, upscaling or content edits. Decoded RGBA pixels match the source rasterization.
MIT License
Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
Copyright (c) 2025-2026 DeerFlow Authors
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Feature correlations from the same demo
Author replay output showing a correlation heatmap for numeric features in the supplied Titanic sample. The matching titanic.csv attachment is required.
Example input
What interesting findings can be made from this dataset?Example output
The matching dataset is bundled at user-data/uploads/titanic.csv beside the linked source thread. These are author replay artifacts, with no rerun or independent statistical verification by this site.License
Exact complete human question from the pinned public thread, paired with that same thread’s committed output image and CSV attachment. The trace uses Python/matplotlib/seaborn; it is not an image-model generation. No model reasoning text is republished. The trace does not establish invocation of a named data-analysis Skill, so this example is bound only to the accepted DeerFlow Product. Original author PNG normalized to lossless WebP at its native dimensions. No cropping, upscaling or content edits. Decoded RGBA pixels match the source rasterization.
MIT License
Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
Copyright (c) 2025-2026 DeerFlow Authors
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Choose context from the mobile composer
Author interface preview of a picker for Skills, conversations and project files.
License
Published interface screenshot from the original repository. Its configuration and sample content are author-provided; no live integration or source instruction was executed. Original author PNG normalized to lossless WebP at its native dimensions. No cropping, upscaling or content edits. Decoded RGBA pixels match the source rasterization.
MIT License
Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
Copyright (c) 2025-2026 DeerFlow Authors
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Inspect an installed extension
Author interface preview of the Capability Center and its Bookmarks extension, with English and Chinese interface text.
License
Published interface screenshot from the original repository. Its configuration and sample content are author-provided; no live integration or source instruction was executed. Original author PNG normalized to lossless WebP at its native dimensions. No cropping, upscaling or content edits. Decoded RGBA pixels match the source rasterization.
MIT License
Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
Copyright (c) 2025-2026 DeerFlow Authors
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Product overview
DeerFlow is an open-source agent workspace for research, coding and content tasks, combining tools, reusable Skills and sandboxed workspaces.
What DeerFlow offers
- 01Coordinate multi-step tasks with tools and delegated agents
- 02Load and extend reusable Skills
- 03Work with files in configured sandbox workspaces
- 04Keep task context and memory across longer workflows
What makes it different
The reviewed 2.0 harness coordinates models, tools and workspaces around longer tasks. The operator configures model access and sandbox services; the legacy 1.x branch has a different implementation.
Who should consider it
- Developers configuring research and coding agents
- Technical users running multi-step content workflows
What to verify
- Try one real task and compare its output with your requirements.
- Check what data the service stores, shares, or uses for training.
- Confirm current pricing, usage limits, and cancellation terms on the official site.
Official link checked on Oct 22, 2026. Feature claims have not been independently tested.
