Restore gallery and real-world images and site badges

README.md

@@ -1,10 +1,12 @@
 <p align="center">
-  <a href="https://github.com/openrobokit/graspkit"><img src="https://raw.githubusercontent.com/openrobokit/graspkit/main/docs/assets/graspkit-banner.webp" alt="GraspKit" width="100%"></a>
+  <a href="https://github.com/openrobokit/graspkit"><img src="https://openrobokit.com/assets/docs/graspkit/assets/graspkit-banner.webp" alt="GraspKit" width="100%"></a>
 </p>
 
 <p align="center">
   <a href="https://github.com/openrobokit/graspkit/blob/main/pyproject.toml"><img src="https://img.shields.io/badge/python-3.10%2B-blue" alt="Python 3.10+"></a>
   <a href="https://github.com/openrobokit/graspkit/actions/workflows/build.yaml"><img src="https://img.shields.io/badge/CI-GitHub_Actions-blue" alt="GitHub Actions"></a>
+  <a href="https://openrobokit.com/graspkit/"><img src="https://img.shields.io/badge/homepage-GraspKit-blue" alt="Homepage"></a>
+  <a href="https://openrobokit.com/docs/graspkit/overview/"><img src="https://img.shields.io/badge/docs-GraspKit-blue" alt="Docs"></a>
   <a href="https://github.com/openrobokit/graspkit/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue" alt="MIT License"></a>
 </p>
 
@@ -13,8 +15,14 @@ and dexterous hands.
 It synthesizes grasps, plans trajectories, trains policies, and deploys them
 in simulation or on real robots; it is built on top of [RoboKit](https://github.com/openrobokit/robokit).
 
+<p align="center">
+  <a href="https://openrobokit.com/graspkit/"><img src="https://openrobokit.com/assets/docs/graspkit/assets/showcase/real-world.webp" alt="Robot hands grasping objects during a real-world run" width="100%"></a><br>
+  <em>Real-world grasping with dexterous robot hands in San Diego.</em>
+</p>
+
 ## Quick links
 
+- [Homepage](https://openrobokit.com/graspkit/)
 - [Docs](https://github.com/openrobokit/graspkit/blob/main/docs/guide/overview.md)
 - [Quickstart](https://github.com/openrobokit/graspkit/blob/main/docs/guide/quickstart.md)
 - [Installation](https://github.com/openrobokit/graspkit/blob/main/docs/guide/installation.md)
@@ -35,21 +43,21 @@ See [Installation](https://github.com/openrobokit/graspkit/blob/main/docs/guide/
 <table width="100%">
   <tr>
     <td width="50%">
-      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/grasp_optim.md"><img src="https://raw.githubusercontent.com/openrobokit/graspkit/main/docs/assets/features/grasp-optimization.webp" alt="1,024 dexterous grasps optimized in one GPU batch" width="100%"></a><br>
+      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/grasp_optim.md"><img src="https://openrobokit.com/assets/docs/graspkit/assets/features/grasp-optimization.webp" alt="1,024 dexterous grasps optimized in one GPU batch" width="100%"></a><br>
       <strong><a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/grasp_optim.md">Fast grasp optimization</a></strong>
     </td>
     <td width="50%">
-      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/training.md"><img src="https://raw.githubusercontent.com/openrobokit/graspkit/main/docs/assets/features/policy-learning.webp" alt="Object point cloud mapped by a grasp generator to a hand grasp" width="100%"></a><br>
+      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/training.md"><img src="https://openrobokit.com/assets/docs/graspkit/assets/features/policy-learning.webp" alt="Object point cloud mapped by a grasp generator to a hand grasp" width="100%"></a><br>
       <strong><a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/training.md">Policy learning</a></strong>
     </td>
   </tr>
   <tr>
     <td>
-      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/reference/robot_hands.md"><img src="https://raw.githubusercontent.com/openrobokit/graspkit/main/docs/assets/features/embodiments.webp" alt="Different arms with dexterous hands grasping the same object in simulation" width="100%"></a><br>
+      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/reference/robot_hands.md"><img src="https://openrobokit.com/assets/docs/graspkit/assets/features/embodiments.webp" alt="Different arms with dexterous hands grasping the same object in simulation" width="100%"></a><br>
       <strong><a href="https://github.com/openrobokit/graspkit/blob/main/docs/reference/robot_hands.md">All arms, all hands</a></strong>
     </td>
     <td>
-      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/real_deployment.md"><img src="https://raw.githubusercontent.com/openrobokit/graspkit/main/docs/assets/features/sim-to-real.webp" alt="Two real xArm robots with dexterous hands, diagonally split with their aligned simulation twin" width="100%"></a><br>
+      <a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/real_deployment.md"><img src="https://openrobokit.com/assets/docs/graspkit/assets/features/sim-to-real.webp" alt="Two real xArm robots with dexterous hands, diagonally split with their aligned simulation twin" width="100%"></a><br>
       <strong><a href="https://github.com/openrobokit/graspkit/blob/main/docs/guide/real_deployment.md">Real-world robustness</a></strong>
     </td>
   </tr>

docs/guide/deployment.md

@@ -29,32 +29,32 @@ Click **Capture** to start over whenever it is enabled.
 
 Pick a **Scene**. Click **Capture**. See the camera image on the right.
 
-![Captured simulation RGB alongside the xArm7 and xHand scene](../assets/guide/deploy_capture.webp)
+![Captured simulation RGB alongside the xArm7 and xHand scene](https://openrobokit.com/assets/docs/graspkit/assets/guide/deploy_capture.webp)
 
 ### 2. Segment
 
 Click **Segment**. Click the object in the large image. Blue = selected. Click **Submit Mask**.
 
-![SAM2 object mask with a positive prompt and the segmentation controls](../assets/guide/deploy_segment.webp)
+![SAM2 object mask with a positive prompt and the segmentation controls](https://openrobokit.com/assets/docs/graspkit/assets/guide/deploy_segment.webp)
 
 ### 3. Check Grasp/Trajectory
 
 Click **Check Grasp/Trajectory**. **Grasp** picks a candidate. **Trajectory Step** previews its motion.
 
-![A candidate grasp and its planned trajectory beside the segmented red cloud](../assets/guide/deploy_plan.webp)
+![A candidate grasp and its planned trajectory beside the segmented red cloud](https://openrobokit.com/assets/docs/graspkit/assets/guide/deploy_plan.webp)
 
 ### 4. Deploy
 
 Click **Deploy** to move. **Deploy Until Step** sets where to stop.
 Raise it; click **Deploy** to continue.
 
-![Simulation paused partway through execution using Deploy Until Step](../assets/guide/deploy_pause.webp)
+![Simulation paused partway through execution using Deploy Until Step](https://openrobokit.com/assets/docs/graspkit/assets/guide/deploy_pause.webp)
 
 ### 5. Release
 
 Click **Release**. Hand opens. Arm goes home. `--loop` lets you repeat.
 
-![The completed simulated trajectory, ready to release and park](../assets/guide/deploy_release.webp)
+![The completed simulated trajectory, ready to release and park](https://openrobokit.com/assets/docs/graspkit/assets/guide/deploy_release.webp)
 
 ### Controls and options
 

docs/guide/grasp_optim.md

@@ -1,7 +1,7 @@
 # Grasp Optimization (`grasp_optim`)
 
 
-| ![A floating Sharpa hand optimizing a grasp on a cube](../assets/guide/sharpa_optim.webp) | ![An xArm7 + Sharpa optimizing arm and hand joints together](../assets/guide/xarm7_sharpa_optim.webp) |
+| ![A floating Sharpa hand optimizing a grasp on a cube](https://openrobokit.com/assets/docs/graspkit/assets/guide/sharpa_optim.webp) | ![An xArm7 + Sharpa optimizing arm and hand joints together](https://openrobokit.com/assets/docs/graspkit/assets/guide/xarm7_sharpa_optim.webp) |
 | ----------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------- |
 | A floating hand optimizing a grasp.                                                       | An arm and hand optimized together on a fixed base.                                                   |
 
@@ -67,7 +67,7 @@ The sections below build the same stage step by step in Python.
 
 ### Sampling initial poses
 
-![Initial Sharpa hand poses sampled around a mustard bottle](../assets/guide/sharpa_init.webp)
+![Initial Sharpa hand poses sampled around a mustard bottle](https://openrobokit.com/assets/docs/graspkit/assets/guide/sharpa_init.webp)
 
 Init samples drawn with the palm-down cone below: red is the hand axis (`palm_forward`), gray is the target
 (world -z). Both start at the wrist; the label is their angle, always within the 60 deg cone.
@@ -118,7 +118,7 @@ grasp_qpos = optimizer.optimize([scene], num_samples_per_scene=8)  # (1, 8, 18)
 Grasp filters live in `GraspTrajBuilderConfig` and run first inside `GraspTrajBuilder.build()`. They
 only mark grasps, they don't drop them. Call `filter_grasps` to run them alone, without motion planning:
 
-![A passing and a failing xHand grasp for each filter: contact-avoid, penetration, hand axis](../assets/guide/grasp_filters.png)
+![A passing and a failing xHand grasp for each filter: contact-avoid, penetration, hand axis](https://openrobokit.com/assets/docs/graspkit/assets/guide/grasp_filters.png)
 
 ```python
 from graspkit.configs.experiment_config import GraspTrajBuilderConfig

docs/guide/overview.md

@@ -1,10 +1,14 @@
-![GraspKit](../assets/graspkit-banner.webp)
+![GraspKit](https://openrobokit.com/assets/docs/graspkit/assets/graspkit-banner.webp)
 
 GraspKit is an end-to-end toolkit for manipulation across various robot embodiments, including grippers
 and dexterous hands.
 It synthesizes grasps, plans trajectories, trains policies, and deploys them
 in simulation or on real robots; it is built on top of [RoboKit](https://github.com/openrobokit/robokit).
 
+[![Robot hands grasping objects during a real-world run](https://openrobokit.com/assets/docs/graspkit/assets/showcase/real-world.webp)](https://openrobokit.com/graspkit/)
+
+*Real-world grasping with dexterous robot hands in San Diego.*
+
 ## Key features
 
 - **Fast grasp optimization.** GPU optimizers refine thousands of grasp seeds per batch against
@@ -18,10 +22,10 @@ in simulation or on real robots; it is built on top of [RoboKit](https://github.
 - **Real-world validation.** Run trained policies on xArm and Vega robots with calibrated cameras
   and collision-checked motions. See [Real deployment](real_deployment.md).
 
-| [![Fast grasp optimization](../assets/features/grasp-optimization.webp)](grasp_optim.md) | [![Policy training](../assets/features/policy-learning.webp)](training.md) |
+| [![Fast grasp optimization](https://openrobokit.com/assets/docs/graspkit/assets/features/grasp-optimization.webp)](grasp_optim.md) | [![Policy training](https://openrobokit.com/assets/docs/graspkit/assets/features/policy-learning.webp)](training.md) |
 | --- | --- |
 | **[Fast grasp optimization](grasp_optim.md)** | **[Performant policy training](training.md)** |
-| [![All arms, all hands](../assets/features/embodiments.webp)](../reference/robot_hands.md) | [![Real-world robustness](../assets/features/sim-to-real.webp)](real_deployment.md) |
+| [![All arms, all hands](https://openrobokit.com/assets/docs/graspkit/assets/features/embodiments.webp)](../reference/robot_hands.md) | [![Real-world robustness](https://openrobokit.com/assets/docs/graspkit/assets/features/sim-to-real.webp)](real_deployment.md) |
 | **[All arms, all hands](../reference/robot_hands.md)** | **[Real-world robustness](real_deployment.md)** |
 
 ## How it works
@@ -42,3 +46,4 @@ For the graph of stages and the artifacts between them, see
 - [Quickstart](quickstart.md)
 - [Installation](installation.md)
 - [GitHub](https://github.com/openrobokit/graspkit)
+- [Website](https://openrobokit.com/graspkit/)

docs/guide/quickstart.md

@@ -11,6 +11,8 @@ uvx --from graspkit python -m graspkit.gallery.grasp_optim
 Pick a hand, arm, and object. Press *Optimize* to generate grasps, then browse them with the
 *Grasp* slider. The gallery caches results for the next visit.
 
+![The grasp gallery switching from xHand to a Robotiq gripper](https://openrobokit.com/assets/docs/graspkit/assets/gallery/grasp_optim.webp)
+
 ## Optimize one grasp
 
 After [installing](installation.md), run the smallest grasp optimization example:

scripts/benchmark/grasp_opt/bench_grasp_optim_results.robokit_opt.md

@@ -1,4 +1,4 @@
-# Grasp Benchmark Results — mesh_vs_low_sdf (graspkit on robokit contact var and solvers; volcan RTX 3090, idle GPU)
+# Grasp Benchmark Results — mesh_vs_low_sdf (graspkit on robokit contact var and solvers; RTX 3090, idle GPU)
 Date: 2026-09-05 01:29
 ## xhand_right_dev_mesh_vs_low_sdf (floating task, 8 scenes × 64 samples = 512 grasps)
 | Metric | mesh_warp_gd | mesh_warp_lm | low_sdf_warp_gd | low_sdf_warp_lm |