From 66ee081dc181fc731994f50bb99b0a52a2761310 Mon Sep 17 00:00:00 2001 From: Teknium <127238744+teknium1@users.noreply.github.com> Date: Sun, 19 Apr 2026 05:14:17 -0700 Subject: [PATCH] skills: move 7 niche mlops/mcp skills to optional (#12474) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Built-in → optional-skills/: mlops/training/peft → optional-skills/mlops/peft mlops/training/pytorch-fsdp → optional-skills/mlops/pytorch-fsdp mlops/models/clip → optional-skills/mlops/clip mlops/models/stable-diffusion → optional-skills/mlops/stable-diffusion mlops/models/whisper → optional-skills/mlops/whisper mlops/cloud/modal → optional-skills/mlops/modal mcp/mcporter → optional-skills/mcp/mcporter Built-in mlops training kept: axolotl, trl-fine-tuning, unsloth. Built-in mlops models kept: audiocraft, segment-anything. Built-in mlops evaluation/research/huggingface-hub/inference all kept. native-mcp stays built-in (documents the native MCP tool); mcporter was a redundant alternative CLI. Also: removed now-empty skills/mlops/cloud/ dir, refreshed skills/mlops/models/DESCRIPTION.md and skills/mcp/DESCRIPTION.md to match what's left, and synchronized both catalog pages (skills-catalog.md, optional-skills-catalog.md). --- .../mcp/mcporter/SKILL.md | 0 .../mlops}/clip/SKILL.md | 0 .../mlops}/clip/references/applications.md | 0 .../mlops}/modal/SKILL.md | 0 .../mlops}/modal/references/advanced-usage.md | 0 .../mlops}/modal/references/troubleshooting.md | 0 .../mlops}/peft/SKILL.md | 0 .../mlops}/peft/references/advanced-usage.md | 0 .../mlops}/peft/references/troubleshooting.md | 0 .../mlops}/pytorch-fsdp/SKILL.md | 0 .../mlops}/pytorch-fsdp/references/index.md | 0 .../mlops}/pytorch-fsdp/references/other.md | 0 .../mlops}/stable-diffusion/SKILL.md | 0 .../references/advanced-usage.md | 0 .../references/troubleshooting.md | 0 .../mlops}/whisper/SKILL.md | 0 .../mlops}/whisper/references/languages.md | 0 skills/mcp/DESCRIPTION.md | 2 +- skills/mlops/cloud/DESCRIPTION.md | 3 --- skills/mlops/models/DESCRIPTION.md | 2 +- .../docs/reference/optional-skills-catalog.md | 7 +++++++ website/docs/reference/skills-catalog.md | 16 +--------------- 22 files changed, 10 insertions(+), 20 deletions(-) rename {skills => optional-skills}/mcp/mcporter/SKILL.md (100%) rename {skills/mlops/models => optional-skills/mlops}/clip/SKILL.md (100%) rename {skills/mlops/models => optional-skills/mlops}/clip/references/applications.md (100%) rename {skills/mlops/cloud => optional-skills/mlops}/modal/SKILL.md (100%) rename {skills/mlops/cloud => optional-skills/mlops}/modal/references/advanced-usage.md (100%) rename {skills/mlops/cloud => optional-skills/mlops}/modal/references/troubleshooting.md (100%) rename {skills/mlops/training => optional-skills/mlops}/peft/SKILL.md (100%) rename {skills/mlops/training => optional-skills/mlops}/peft/references/advanced-usage.md (100%) rename {skills/mlops/training => optional-skills/mlops}/peft/references/troubleshooting.md (100%) rename {skills/mlops/training => optional-skills/mlops}/pytorch-fsdp/SKILL.md (100%) rename {skills/mlops/training => optional-skills/mlops}/pytorch-fsdp/references/index.md (100%) rename {skills/mlops/training => optional-skills/mlops}/pytorch-fsdp/references/other.md (100%) rename {skills/mlops/models => optional-skills/mlops}/stable-diffusion/SKILL.md (100%) rename {skills/mlops/models => optional-skills/mlops}/stable-diffusion/references/advanced-usage.md (100%) rename {skills/mlops/models => optional-skills/mlops}/stable-diffusion/references/troubleshooting.md (100%) rename {skills/mlops/models => optional-skills/mlops}/whisper/SKILL.md (100%) rename {skills/mlops/models => optional-skills/mlops}/whisper/references/languages.md (100%) delete mode 100644 skills/mlops/cloud/DESCRIPTION.md diff --git a/skills/mcp/mcporter/SKILL.md b/optional-skills/mcp/mcporter/SKILL.md similarity index 100% rename from skills/mcp/mcporter/SKILL.md rename to optional-skills/mcp/mcporter/SKILL.md diff --git a/skills/mlops/models/clip/SKILL.md b/optional-skills/mlops/clip/SKILL.md similarity index 100% rename from skills/mlops/models/clip/SKILL.md rename to optional-skills/mlops/clip/SKILL.md diff --git a/skills/mlops/models/clip/references/applications.md b/optional-skills/mlops/clip/references/applications.md similarity index 100% rename from skills/mlops/models/clip/references/applications.md rename to optional-skills/mlops/clip/references/applications.md diff --git a/skills/mlops/cloud/modal/SKILL.md b/optional-skills/mlops/modal/SKILL.md similarity index 100% rename from skills/mlops/cloud/modal/SKILL.md rename to optional-skills/mlops/modal/SKILL.md diff --git a/skills/mlops/cloud/modal/references/advanced-usage.md b/optional-skills/mlops/modal/references/advanced-usage.md similarity index 100% rename from skills/mlops/cloud/modal/references/advanced-usage.md rename to optional-skills/mlops/modal/references/advanced-usage.md diff --git a/skills/mlops/cloud/modal/references/troubleshooting.md b/optional-skills/mlops/modal/references/troubleshooting.md similarity index 100% rename from skills/mlops/cloud/modal/references/troubleshooting.md rename to optional-skills/mlops/modal/references/troubleshooting.md diff --git a/skills/mlops/training/peft/SKILL.md b/optional-skills/mlops/peft/SKILL.md similarity index 100% rename from skills/mlops/training/peft/SKILL.md rename to optional-skills/mlops/peft/SKILL.md diff --git a/skills/mlops/training/peft/references/advanced-usage.md b/optional-skills/mlops/peft/references/advanced-usage.md similarity index 100% rename from skills/mlops/training/peft/references/advanced-usage.md rename to optional-skills/mlops/peft/references/advanced-usage.md diff --git a/skills/mlops/training/peft/references/troubleshooting.md b/optional-skills/mlops/peft/references/troubleshooting.md similarity index 100% rename from skills/mlops/training/peft/references/troubleshooting.md rename to optional-skills/mlops/peft/references/troubleshooting.md diff --git a/skills/mlops/training/pytorch-fsdp/SKILL.md b/optional-skills/mlops/pytorch-fsdp/SKILL.md similarity index 100% rename from skills/mlops/training/pytorch-fsdp/SKILL.md rename to optional-skills/mlops/pytorch-fsdp/SKILL.md diff --git a/skills/mlops/training/pytorch-fsdp/references/index.md b/optional-skills/mlops/pytorch-fsdp/references/index.md similarity index 100% rename from skills/mlops/training/pytorch-fsdp/references/index.md rename to optional-skills/mlops/pytorch-fsdp/references/index.md diff --git a/skills/mlops/training/pytorch-fsdp/references/other.md b/optional-skills/mlops/pytorch-fsdp/references/other.md similarity index 100% rename from skills/mlops/training/pytorch-fsdp/references/other.md rename to optional-skills/mlops/pytorch-fsdp/references/other.md diff --git a/skills/mlops/models/stable-diffusion/SKILL.md b/optional-skills/mlops/stable-diffusion/SKILL.md similarity index 100% rename from skills/mlops/models/stable-diffusion/SKILL.md rename to optional-skills/mlops/stable-diffusion/SKILL.md diff --git a/skills/mlops/models/stable-diffusion/references/advanced-usage.md b/optional-skills/mlops/stable-diffusion/references/advanced-usage.md similarity index 100% rename from skills/mlops/models/stable-diffusion/references/advanced-usage.md rename to optional-skills/mlops/stable-diffusion/references/advanced-usage.md diff --git a/skills/mlops/models/stable-diffusion/references/troubleshooting.md b/optional-skills/mlops/stable-diffusion/references/troubleshooting.md similarity index 100% rename from skills/mlops/models/stable-diffusion/references/troubleshooting.md rename to optional-skills/mlops/stable-diffusion/references/troubleshooting.md diff --git a/skills/mlops/models/whisper/SKILL.md b/optional-skills/mlops/whisper/SKILL.md similarity index 100% rename from skills/mlops/models/whisper/SKILL.md rename to optional-skills/mlops/whisper/SKILL.md diff --git a/skills/mlops/models/whisper/references/languages.md b/optional-skills/mlops/whisper/references/languages.md similarity index 100% rename from skills/mlops/models/whisper/references/languages.md rename to optional-skills/mlops/whisper/references/languages.md diff --git a/skills/mcp/DESCRIPTION.md b/skills/mcp/DESCRIPTION.md index 627c20ea1..30a066033 100644 --- a/skills/mcp/DESCRIPTION.md +++ b/skills/mcp/DESCRIPTION.md @@ -1,3 +1,3 @@ --- -description: Skills for working with MCP (Model Context Protocol) servers, tools, and integrations. Includes the built-in native MCP client (configure servers in config.yaml for automatic tool discovery) and the mcporter CLI bridge for ad-hoc server interaction. +description: Skills for working with MCP (Model Context Protocol) servers, tools, and integrations. Documents the built-in native MCP client — configure servers in config.yaml for automatic tool discovery. --- diff --git a/skills/mlops/cloud/DESCRIPTION.md b/skills/mlops/cloud/DESCRIPTION.md deleted file mode 100644 index 32675823e..000000000 --- a/skills/mlops/cloud/DESCRIPTION.md +++ /dev/null @@ -1,3 +0,0 @@ ---- -description: GPU cloud providers and serverless compute platforms for ML workloads. ---- diff --git a/skills/mlops/models/DESCRIPTION.md b/skills/mlops/models/DESCRIPTION.md index 8170b517f..8f7e66956 100644 --- a/skills/mlops/models/DESCRIPTION.md +++ b/skills/mlops/models/DESCRIPTION.md @@ -1,3 +1,3 @@ --- -description: Specific model architectures and tools — computer vision (CLIP, SAM, Stable Diffusion), speech (Whisper), audio generation (AudioCraft), and multimodal models (LLaVA). +description: Specific model architectures and tools — image segmentation (Segment Anything / SAM) and audio generation (AudioCraft / MusicGen). Additional model skills (CLIP, Stable Diffusion, Whisper, LLaVA) are available as optional skills. --- diff --git a/website/docs/reference/optional-skills-catalog.md b/website/docs/reference/optional-skills-catalog.md index 044060e9d..f5dd2ac5b 100644 --- a/website/docs/reference/optional-skills-catalog.md +++ b/website/docs/reference/optional-skills-catalog.md @@ -83,6 +83,7 @@ hermes skills uninstall | Skill | Description | |-------|-------------| | **fastmcp** | Build, test, inspect, install, and deploy MCP servers with FastMCP in Python. Covers wrapping APIs or databases as MCP tools, exposing resources or prompts, and deployment. | +| **mcporter** | The `mcporter` CLI — list, configure, auth, and call MCP servers/tools directly (HTTP or stdio) from the terminal. Useful for ad-hoc MCP interactions; for always-on tool discovery use the built-in `native-mcp` client instead. | ## Migration @@ -98,6 +99,7 @@ The largest optional category — covers the full ML pipeline from data curation |-------|-------------| | **accelerate** | Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for DeepSpeed/FSDP/Megatron/DDP. | | **chroma** | Open-source embedding database. Store embeddings and metadata, perform vector and full-text search. Simple 4-function API for RAG and semantic search. | +| **clip** | OpenAI's vision-language model connecting images and text. Zero-shot image classification, image-text matching, and cross-modal retrieval. Trained on 400M image-text pairs. Use for image search, content moderation, or vision-language tasks without fine-tuning. | | **faiss** | Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). | | **flash-attention** | Optimize transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Supports PyTorch SDPA, flash-attn library, H100 FP8, and sliding window. | | **guidance** | Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance — Microsoft Research's constrained generation framework. | @@ -106,15 +108,20 @@ The largest optional category — covers the full ML pipeline from data curation | **instructor** | Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, and stream partial results. | | **lambda-labs** | Reserved and on-demand GPU cloud instances for ML training and inference. SSH access, persistent filesystems, and multi-node clusters. | | **llava** | Large Language and Vision Assistant — visual instruction tuning and image-based conversations combining CLIP vision with LLaMA language models. | +| **modal** | Serverless GPU cloud platform for running ML workloads. On-demand GPU access without infrastructure management, ML model deployment as APIs, or batch jobs with automatic scaling. | | **nemo-curator** | GPU-accelerated data curation for LLM training. Fuzzy deduplication (16x faster), quality filtering (30+ heuristics), semantic dedup, PII redaction. Scales with RAPIDS. | +| **peft-fine-tuning** | Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Train <1% of parameters with minimal accuracy loss for 7B–70B models on limited GPU memory. HuggingFace's official PEFT library. | | **pinecone** | Managed vector database for production AI. Auto-scaling, hybrid search (dense + sparse), metadata filtering, and low latency (under 100ms p95). | +| **pytorch-fsdp** | Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP — parameter sharding, mixed precision, CPU offloading, FSDP2. | | **pytorch-lightning** | High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks, and minimal boilerplate. | | **qdrant** | High-performance vector similarity search engine. Rust-powered with fast nearest neighbor search, hybrid search with filtering, and scalable vector storage. | | **saelens** | Train and analyze Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. | | **simpo** | Simple Preference Optimization — reference-free alternative to DPO with better performance (+6.4 pts on AlpacaEval 2.0). No reference model needed. | | **slime** | LLM post-training with RL using Megatron+SGLang framework. Custom data generation workflows and tight Megatron-LM integration for RL scaling. | +| **stable-diffusion-image-generation** | State-of-the-art text-to-image generation with Stable Diffusion via HuggingFace Diffusers. Text-to-image, image-to-image translation, inpainting, and custom diffusion pipelines. | | **tensorrt-llm** | Optimize LLM inference with NVIDIA TensorRT for maximum throughput. 10-100x faster than PyTorch on A100/H100 with quantization (FP8/INT4) and in-flight batching. | | **torchtitan** | PyTorch-native distributed LLM pretraining with 4D parallelism (FSDP2, TP, PP, CP). Scale from 8 to 512+ GPUs with Float8 and torch.compile. | +| **whisper** | OpenAI's general-purpose speech recognition. 99 languages, transcription, translation to English, and language ID. Six model sizes from tiny (39M) to large (1550M). Best for robust multilingual ASR. | ## Productivity diff --git a/website/docs/reference/skills-catalog.md b/website/docs/reference/skills-catalog.md index 16be6a658..ffe489d36 100644 --- a/website/docs/reference/skills-catalog.md +++ b/website/docs/reference/skills-catalog.md @@ -114,7 +114,6 @@ Skills for working with MCP (Model Context Protocol) servers, tools, and integra | Skill | Description | Path | |-------|-------------|------| -| `mcporter` | Use the mcporter CLI to list, configure, auth, and call MCP servers/tools directly (HTTP or stdio), including ad-hoc servers, config edits, and CLI/type generation. | `mcp/mcporter` | | `native-mcp` | Built-in MCP (Model Context Protocol) client that connects to external MCP servers, discovers their tools, and registers them as native Hermes Agent tools. Supports stdio and HTTP transports with automatic reconnection, security filtering, and zero-config tool injection. | `mcp/native-mcp` | ## media @@ -136,14 +135,6 @@ General-purpose ML operations tools — model hub management, dataset operations |-------|-------------|------| | `huggingface-hub` | Hugging Face Hub CLI (hf) — search, download, and upload models and datasets, manage repos, query datasets with SQL, deploy inference endpoints, manage Spaces and buckets. | `mlops/huggingface-hub` | -## mlops/cloud - -GPU cloud providers and serverless compute platforms for ML workloads. - -| Skill | Description | Path | -|-------|-------------|------| -| `modal-serverless-gpu` | Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling. | `mlops/cloud/modal` | - ## mlops/evaluation Model evaluation benchmarks, experiment tracking, and interpretability tools. @@ -166,15 +157,12 @@ Model serving, quantization (GGUF/GPTQ), structured output, inference optimizati ## mlops/models -Specific model architectures — computer vision (CLIP, SAM, Stable Diffusion), speech (Whisper), and audio generation (AudioCraft). +Specific model architectures — image segmentation (SAM) and audio generation (AudioCraft / MusicGen). Additional model skills (CLIP, Stable Diffusion, Whisper, LLaVA) are available as optional skills. | Skill | Description | Path | |-------|-------------|------| | `audiocraft-audio-generation` | PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation. | `mlops/models/audiocraft` | -| `clip` | OpenAI's model connecting vision and language. Enables zero-shot image classification, image-text matching, and cross-modal retrieval. Trained on 400M image-text pairs. Use for image search, content moderation, or vision-language tasks without fine-tuning. Best for general-pur… | `mlops/models/clip` | | `segment-anything-model` | 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. | `mlops/models/segment-anything` | -| `stable-diffusion-image-generation` | 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. | `mlops/models/stable-diffusion` | -| `whisper` | OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio pr… | `mlops/models/whisper` | ## mlops/research @@ -192,8 +180,6 @@ Fine-tuning, RLHF/DPO/GRPO training, distributed training frameworks, and optimi |-------|-------------|------| | `axolotl` | Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support | `mlops/training/axolotl` | | `fine-tuning-with-trl` | Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace … | `mlops/training/trl-fine-tuning` | -| `peft-fine-tuning` | Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library… | `mlops/training/peft` | -| `pytorch-fsdp` | Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2 | `mlops/training/pytorch-fsdp` | | `unsloth` | Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization | `mlops/training/unsloth` | ## note-taking