DexForce · yuecideng · Mar 11, 2026 · Mar 11, 2026 · Mar 11, 2026
diff --git a/AGENTS.md b/AGENTS.md
@@ -207,7 +207,17 @@ Include:
 
 ### Adding a New Robot
 
-Refer to `embodichain/lab/sim/robots/` for existing examples (`dexforce_w1`, `cobotmagic`). Each robot needs: a config (`cfg.py`), parameters (`params.py`), types (`types.py`), and utilities (`utils.py`).
+Refer to `docs/source/tutorial/add_robot.rst` for a detailed guide. The basic structure requires:
+
+- A config class (inheriting from `RobotCfg`)
+- URDF configuration for the robot
+- Control parts definition
+- IK solver configuration
+- Drive properties for joint physics
+
+For complex robots with multiple variants (like `dexforce_w1`), use a package structure with `types.py`, `params.py`, `utils.py`, and `cfg.py`.
+
+Also add robot documentation in `docs/source/resources/robot/` (see existing examples: `cobotmagic.md`, `dexforce_w1.md`) and update `docs/source/resources/robot/index.rst` to include the new robot.
 
 ### Adding a New Task Environment
 

diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rl.algo.rst b/docs/source/api_reference/embodichain/embodichain.agents.rl.algo.rst
@@ -0,0 +1,13 @@
+embodichain.agents.rl.algo
+==========================
+
+.. automodule:: embodichain.agents.rl.algo
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      build_algo
+      get_registered_algo_names
+
diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rl.buffer.rst b/docs/source/api_reference/embodichain/embodichain.agents.rl.buffer.rst
@@ -0,0 +1,6 @@
+embodichain.agents.rl.buffer
+============================
+
+.. automodule:: embodichain.agents.rl.buffer
+
+
diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rl.models.rst b/docs/source/api_reference/embodichain/embodichain.agents.rl.models.rst
@@ -0,0 +1,16 @@
+embodichain.agents.rl.models
+============================
+
+.. automodule:: embodichain.agents.rl.models
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      build_mlp_from_cfg
+      build_policy
+      get_policy_class
+      get_registered_policy_names
+      register_policy
+
diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rl.rst b/docs/source/api_reference/embodichain/embodichain.agents.rl.rst
@@ -0,0 +1,55 @@
+embodichain.agents.rl
+=======================
+
+.. automodule:: embodichain.agents.rl
+
+   .. rubric:: Submodules
+
+   .. autosummary::
+      :toctree: .
+
+      algo
+      buffer
+      models
+      train
+      utils
+
+Algorithms
+----------
+
+.. automodule:: embodichain.agents.rl.algo
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Rollout Buffer
+--------------
+
+.. automodule:: embodichain.agents.rl.buffer
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Policy Models
+-------------
+
+.. automodule:: embodichain.agents.rl.models
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Training
+--------
+
+.. automodule:: embodichain.agents.rl.train
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Utilities
+---------
+
+.. automodule:: embodichain.agents.rl.utils
+   :members:
+   :undoc-members:
+   :show-inheritance:
diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rl.train.rst b/docs/source/api_reference/embodichain/embodichain.agents.rl.train.rst
@@ -0,0 +1,14 @@
+embodichain.agents.rl.train
+===========================
+
+.. automodule:: embodichain.agents.rl.train
+
+
+   .. rubric:: Functions
+
+   .. autosummary::
+
+      main
+      parse_args
+      train_from_config
+
diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rl.utils.rst b/docs/source/api_reference/embodichain/embodichain.agents.rl.utils.rst
@@ -0,0 +1,6 @@
+embodichain.agents.rl.utils
+===========================
+
+.. automodule:: embodichain.agents.rl.utils
+
+
diff --git a/docs/source/api_reference/embodichain/embodichain.agents.rst b/docs/source/api_reference/embodichain/embodichain.agents.rst
@@ -51,44 +51,3 @@ Reinforcement Learning
       models
       train
       utils
-
-Algorithms
-----------
-
-.. automodule:: embodichain.agents.rl.algo
-   :members:
-   :undoc-members:
-   :show-inheritance:
-
-Rollout Buffer
---------------
-
-.. automodule:: embodichain.agents.rl.buffer
-   :members:
-   :undoc-members:
-   :show-inheritance:
-
-Policy Models
-------------
-
-.. automodule:: embodichain.agents.rl.models
-   :members:
-   :undoc-members:
-   :show-inheritance:
-
-Training
---------
-
-.. automodule:: embodichain.agents.rl.train
-   :members:
-   :undoc-members:
-   :show-inheritance:
-
-Utilities
---------
-
-.. automodule:: embodichain.agents.rl.utils
-   :members:
-   :undoc-members:
-   :show-inheritance:
-
diff --git a/docs/source/overview/agents/online_data.md → docs/source/features/online_data.md b/docs/source/overview/agents/online_data.md → docs/source/features/online_data.md
diff --git a/...esources/toolkits/convex_decomposition.md → ...features/toolkits/convex_decomposition.md b/...esources/toolkits/convex_decomposition.md → ...features/toolkits/convex_decomposition.md
diff --git a/docs/source/resources/toolkits/index.rst → docs/source/features/toolkits/index.rst b/docs/source/resources/toolkits/index.rst → docs/source/features/toolkits/index.rst