Browse skills

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis.

Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging norm stats and GPU memory issues.

Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image.

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases, interpolation methods, and extrapolation strategies for LLMs.

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.

Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn. Use when creating any figure for a conference paper.

Write publication-ready ML/AI/Systems papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM, OSDI, NSDI, ASPLOS, SOSP. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.

Guides researchers through structured ideation frameworks to discover high-impact research directions. Use when exploring new problem spaces, pivoting between projects, or seeking novel angles on existing work.

Applies cognitive science frameworks for creative thinking to CS and AI research ideation. Use when seeking genuinely novel research directions by leveraging combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded creative strategies.

Use when a task spans multiple AEC technologies (Blender, IfcOpenShell, Bonsai, Sverchok) and you need to decide which tool handles which step, or when orchestrating multi-step BIM pipelines such as IFC creation to visualization to authoring. Prevents the common mistake of mixing IfcStore.get_file() with ifcopenshell.open() in the same context, or calling Bonsai operators outside a Bonsai-loaded session. Provides decision trees for technology selection, workflow sequencing, and bridge patterns between tools. Keywords: cross-technology, workflow orchestration, BIM pipeline, technology selection, Blender IfcOpenShell Bonsai integration, multi-step AEC, bridge pattern, which tool should I use, combine Blender and IFC, multi-tool workflow, how to connect tools.

Use when implementing end-to-end BIM workflows that combine IfcOpenShell, Bonsai, and Blender -- such as IFC creation from scratch, model enrichment, validation pipelines, geometry extraction, or batch processing of building models. Prevents the common mistake of skipping unit and context setup before creating geometry, or directly modifying IFC attributes instead of using ifcopenshell.api.run(). Covers property set management across tools, spatial hierarchy patterns, and version compatibility for IFC2X3/IFC4/IFC4X3. Keywords: BIM workflow, IFC creation, model validation, property extraction, batch processing, spatial hierarchy, cross-technology, IfcOpenShell Bonsai Blender, pset management, how to create IFC model, step by step BIM, get properties from IFC.

Use when reviewing, validating, or auditing Blender Python code for correctness. Runs systematic checks for deprecated API usage, context errors, version compatibility issues, threading violations, data reference invalidation, incorrect operator calls, and addon structure compliance. Prevents shipping code with silent version-dependent failures. Keywords: code review, validation, audit, deprecated API, context error, version compatibility, threading, addon structure, code quality, Blender Python, my addon has errors, check my code, why does my script fail.

Use when migrating, porting, or upgrading Blender Python scripts and addons across major versions (3.x to 4.x to 5.x). Provides a systematic migration process covering API renames, removed functions, changed parameters, extension system migration, BGL to GPU module conversion, and bone collection migration. Prevents incomplete migrations that compile but fail at runtime. Keywords: migration, porting, upgrade, version, 3.x to 4.x, 4.x to 5.x, API rename, bgl to gpu, extension migration, bl_info to manifest, bone collection, script stopped working after update, upgrade addon to new Blender.

Use when writing Blender Python scripts that access bpy module structure, RNA data, the context system, dependency graph, or operator invocation. Prevents the common mistake of accessing context attributes outside their valid scope or calling operators without checking poll(). Covers bpy.data, bpy.context, bpy.ops, depsgraph evaluation, and RNA property access patterns across Blender 3.x/4.x/5.x. Keywords: bpy, Blender Python, RNA, context, depsgraph, operator, bpy.data, bpy.ops, how to start scripting Blender, Blender Python basics, access objects in scene, get selected object.