Nous Research
Official@nousresearch
Offers specialized infrastructure for model fine-tuning, vector database management, and high-throughput inference deployment for research and development environments.
Agent Skills by Nous Research
Showing 206 vetted skills indexed across 2 GitHub repositories.
yuanbao
Query Yuanbao group info, list members, and send @mentions or direct messages.
parallel-cli
Perform agent-native web search, deep research, and structured data enrichment via Parallel CLI.
qmd
Search local markdown notes and documents using hybrid keyword and semantic retrieval.
drug-discovery
Query ChEMBL, PubChem, OpenFDA, and OpenTargets for pharmaceutical research and drug-likeness analysis.
osint-investigation
Cross-reference public records to investigate entities, sanctions, and financial relationships.
duckduckgo-search
Perform keyless web, news, image, and video searches via DuckDuckGo.
scrapling
Scrape websites with stealth browsing and Cloudflare bypass.
darwinian-evolver
Evolve prompts, regex, SQL, or code via LLM-driven evolutionary search.
bioinformatics
Index and fetch bioinformatics skills for genomics and computational biology tasks.
pinecone-research
Build agent RAG pipelines and long-term memory using Pinecone vector database.
domain-intel
Perform passive OSINT reconnaissance on domains via subdomains, SSL, WHOIS, and DNS.
research-paper-writing
Write ML research papers for NeurIPS, ICML, and ICLR submission.
minecraft-modpack-server
Configure and deploy modded Minecraft servers from CurseForge or Modrinth packs.
pokemon-player
Play Pokemon games via headless emulator with RAM state and vision analysis.
mpp-agent
Pay HTTP 402 APIs via Machine Payments Protocol clients.
stripe-projects
Provision SaaS services and sync credentials via Stripe Projects CLI.
stripe-link-cli
Complete agent purchases via Stripe Link using virtual cards or Shared Payment Tokens.
hermes-s6-container-supervision
Modify and debug s6-overlay services in the Hermes Docker image.
docker-management
Manage Docker containers, images, volumes, networks, and Compose stacks via CLI.
setup-wizard-generator
Generate interactive bash wizards that guide humans through manual setup procedures.
watchers
Poll RSS feeds, JSON APIs, and GitHub repos with watermark-based deduplication.
jupyter-notebook
Execute Python code in a persistent Jupyter kernel for iterative data exploration.
sherlock
Search social networks for accounts matching a given username.
oss-forensics
Investigate GitHub repositories for supply chain attacks and recover deleted commits.
Frequently Asked Questions About Nous Research
FAQPage SchemaWhat specific model optimization tasks are supported?▼
The registry supports fine-tuning via Axolotl and Unsloth, GGUF quantization for CPU/Apple Silicon inference, and sparse autoencoder training for activation analysis. These capabilities enable efficient model adaptation and deployment across diverse hardware constraints.
Which vector database systems are supported for integration?▼
The platform provides native management for Chroma, Qdrant, and Pinecone. These integrations allow for the storage of embeddings, execution of hybrid metadata filtering, and high-performance vector similarity search within your existing data architecture.
What are the prerequisites for deploying these model workloads?▼
Deployment requires access to GPU-enabled environments, such as Lambda Labs or Modal, and familiarity with YAML-based configuration for fine-tuning. Users must also manage authentication tokens for the respective vector database providers and cloud compute instances.