Recipe/aigen trasfer pt (#295)

* aigen recipe addition

* Update title for post-training document

* close various TODOs

* Add Aigen agriculture recipe to all cookbook indexes and fix video embed

Add the new Cosmos Transfer 2.5 agriculture sim2real recipe to the
Latest Updates tables (README.md, docs/index.md), the Cosmos Transfer
case study table, the All Recipes gallery, and the post-training
SUMMARY.md nav. Fix the cotton video embed to use an HTML video tag
instead of markdown image syntax.

Made-with: Cursor

* Fix formatting issues from pre-commit hooks

Auto-fix trailing whitespace, extra blank lines, and missing blank
lines after markdown headings to pass CI lint checks.

Made-with: Cursor

---------

Co-authored-by: Usman Khan <usman@aigen.io>

Author: paularamo

CONTRIBUTING.md

@@ -276,7 +276,7 @@ For a new recipe at `docs/recipes/[category]/[model-name]/[recipe-name]/[recipe-
 
 **Code Review** - All submissions require review (typically within one week). Respond to feedback promptly and keep discussions professional.
 
-Access to third-party tools - Cosmos recipes are a springboard to help educate and demonstrate how best to use the Cosmos models, libraries, and blueprints. If access to a third-party tool is required for the recipe, credits or access to a free tier should be included to let developers use these recipes without paying for another service. 
+Access to third-party tools - Cosmos recipes are a springboard to help educate and demonstrate how best to use the Cosmos models, libraries, and blueprints. If access to a third-party tool is required for the recipe, credits or access to a free tier should be included to let developers use these recipes without paying for another service.
 Contributions are reviewed on a best-effort basis - we can't provide guarantees around when or if content contributions will be reviewed or merged.
 
 ## Developer Certificate of Origin (Optional)

README.md

@@ -13,6 +13,7 @@ A comprehensive guide for working with the **NVIDIA Cosmos ecosystem**—a suite
 
 | **Date** | **Recipe** | **Model** | **Description** |
 |----------|------------|-----------|-----------------|
+| Apr 21 | [Generate Photorealistic Agricultural Images for Robot Perception Training](https://nvidia-cosmos.github.io/cosmos-cookbook/recipes/post_training/transfer2_5/agtec_scenarios_single_view/post_training.html) | Cosmos Transfer 2.5 | Post-train Cosmos Transfer (depth-conditioned) on agricultural fleet video to generate photorealistic synthetic training images for robot perception |
 | Mar 16 | [Cosmos-Reason2 on Jetson Thor for Edge VLM Perception](https://nvidia-cosmos.github.io/cosmos-cookbook/recipes/inference/reason2/intbot_edge_vlm/inference.html) | Cosmos Reason 2 | Deploy Cosmos-Reason2 on Jetson AGX Thor for social robots (IntBot), with FP8 quantization and TensorRT-Edge-LLM optimization |
 | Mar 15 | [Content-Adaptive Video Compression for Cosmos Curator with Beamr CABR](https://nvidia-cosmos.github.io/cosmos-cookbook/recipes/data_curation/cosmos_cabr/cabr_recipe.html) | Cosmos Curator | Recipe for replacing Cosmos Curator's default CPU-based, fixed-bitrate video encoder with Beamr CABR (Content-Adaptive Bitrate) — a GPU-accelerated video optimization and encoding solution |
 | Mar 15 | [Post-Training Cosmos-H-Surgical-Simulator for Surgical Robotics](https://nvidia-cosmos.github.io/cosmos-cookbook/recipes/post_training/predict2_5/surgical_robotics/post_training.html) | Cosmos Predict 2.5 | Fine-tune Cosmos Predict 2.5 as an action-conditioned surgical simulator for policy evaluation and synthetic data generation using the SutureBot dataset |

docs/index.md

@@ -21,6 +21,7 @@ The Cosmos ecosystem supports the complete Physical AI development lifecycle--fr
 
 | **Date** | **Recipe** | **Model** |
 |----------|------------|-----------|
+| Apr 21 | [Generate Photorealistic Agricultural Images for Robot Perception Training](recipes/post_training/transfer2_5/agtec_scenarios_single_view/post_training.md) | Cosmos Transfer 2.5 |
 | Mar 16 | [Cosmos-Reason2 on Jetson Thor for Edge VLM Perception](recipes/inference/reason2/intbot_edge_vlm/inference.md) | Cosmos Reason 2 |
 | Mar 15 | [Content-Adaptive Video Compression for Cosmos Curator with Beamr CABR](recipes/data_curation/cosmos_cabr/cabr_recipe.md) | Cosmos Curator |
 | Mar 15 | [Post-Training Cosmos-H-Surgical-Simulator for Surgical Robotics](recipes/post_training/predict2_5/surgical_robotics/post_training.md) | Cosmos Predict 2.5 |
@@ -92,6 +93,7 @@ The Cosmos Cookbook includes comprehensive use cases demonstrating real-world ap
 | **Inference** | Sim2Real data augmentation for robotics navigation tasks | [X-Mobility Navigation](recipes/inference/transfer1/inference-x-mobility/inference.md) |
 | **Inference** | Synthetic manipulation motion generation for humanoid robots | [GR00T-Mimic](recipes/inference/transfer1/gr00t-mimic/inference.md) |
 | **Training** | ControlNet post-training for spatially-conditioned multiview AV video generation with world scenario maps | [Multiview AV Generation](recipes/post_training/transfer2_5/av_world_scenario_maps/post_training.md) |
+| **Training** | Post-train Cosmos Transfer (depth-conditioned) on agricultural fleet video for photorealistic synthetic imagery | [Agriculture Sim2Real](recipes/post_training/transfer2_5/agtec_scenarios_single_view/post_training.md) |
 
 ### [**Cosmos Reason**](https://github.com/nvidia-cosmos/cosmos-reason1)
 

docs/recipes/all_recipes.md

@@ -222,6 +222,13 @@
     </div>
     <div class="recipe-category-body">
       <div class="recipe-track" data-page-size="6" aria-label="Robotics recipes">
+        <a class="recipe-card" href="./post_training/transfer2_5/agtec_scenarios_single_view/post_training.html">
+          <div class="recipe-media recipe-media--image" aria-hidden="true">
+            <img src="./post_training/transfer2_5/agtec_scenarios_single_view/assets/cosmos-soybean.gif" alt="" loading="lazy" />
+          </div>
+          <div class="recipe-title">Generate Photorealistic Agricultural Images for Robot Perception Training</div>
+          <div class="recipe-tag recipe-tag--post-training">Post-Training</div>
+        </a>
         <a class="recipe-card" href="./inference/reason2/intbot_edge_vlm/inference.html">
           <div class="recipe-media recipe-media--image" aria-hidden="true">
             <img src="./inference/reason2/intbot_edge_vlm/assets/intbot_gtc_2026.jpg" alt="" loading="lazy" />

docs/recipes/post_training/SUMMARY.md

@@ -6,6 +6,7 @@
 - [Predict for Sports Video Generation](./predict2_5/sports/post_training.md)
 - [Predict for Surgical Robotics Simulation](./predict2_5/surgical_robotics/post_training.md)
 - [Transfer Multiview for Autonomous Vehicles](./transfer2_5/av_world_scenario_maps/post_training.md)
+- [Transfer for Agriculture Sim2Real](./transfer2_5/agtec_scenarios_single_view/post_training.md)
 - [Reason for Physical Plausibility Check](./reason1/physical-plausibility-check/post_training.md)
 - [Reason 2 for Physical Plausibility Check](./reason2/physical-plausibility-check/post_training.md)
 - [Reason for Visual Q&A in ITS](./reason1/intelligent-transportation/post_training.md)