OpenCoven
Official@opencoven · United States of America
Workspace for building with agents.
Agent Skills by OpenCoven
Showing 144 vetted skills indexed across 1 GitHub repositories.
OpenCovenmlflow
Track experiments, version models, and deploy MLflow models.
OpenCovenml-training-recipes
Provide PyTorch training recipes for neural network optimization.
OpenCovenray-data
Process large-scale machine learning datasets with distributed Ray Data execution.
OpenCovenweights-and-biases
Manage ML experiments, track metrics, and run hyperparameter sweeps via Weights & Biases.
OpenCovenfaiss
Perform vector similarity search using Facebook's FAISS library.
OpenCovensentence-transformers
Generate sentence embeddings for semantic similarity, clustering, and retrieval tasks.
OpenCovenara-research-manager
Scan conversation history and log research decisions with provenance tags to an ARA directory.
OpenCovenara-compiler
Convert diverse research inputs into Agent-Native Research Artifacts with cognitive and physical layers.
OpenCovenautoresearch
Automate research workflow management, literature review, and experiment tracking.
OpenCovenevaluating-llms-harness
Evaluate large language models across 60+ academic benchmarks.
OpenCovenevaluating-code-models
Evaluate code generation models across benchmarks using pass@k metrics.
OpenCovenara-rigor-reviewer
Score six research rigor dimensions and generate a severity-ranked report.
OpenCovencreative-thinking-for-research
Generate novel CS and AI research directions using cognitive science frameworks.
OpenCovenbrainstorming-research-ideas
Apply structured ideation frameworks to explore and refine research directions.
OpenCovenacademic-plotting
Generate publication-quality figures for machine learning research papers.
OpenCovenpresenting-conference-talks
Generate Beamer LaTeX PDF and PPTX slides from research papers.
OpenCovenml-paper-writing
Guide ML/AI paper writing with LaTeX templates and citation verification.
OpenCovensystems-paper-writing
Generate paragraph-level structural blueprints for systems papers targeting leading conferences.
OpenCovenhuggingface-accelerate
Automate distributed PyTorch training with HuggingFace Accelerate.
OpenCovenimplementing-llms-litgpt
Train and fine-tune LLMs with LitGPT using LoRA/QLoRA.
OpenCovengguf-quantization
Convert HuggingFace models to GGUF format with 2-8 bit quantization.
OpenCovenserving-llms-vllm
Serve LLMs with OpenAI-compatible endpoints, quantization, and tensor parallelism.
OpenCovenpeft-fine-tuning
Fine-tune large language models with LoRA and QLoRA using PEFT libraries.
OpenCovenllama-cpp
Run large language model inference on non-NVIDIA hardware with GGUF quantization.
Frequently Asked Questions About OpenCoven
FAQPage SchemaWhat specific tasks can I perform using OpenCoven?▼
You can manage research projects, perform GGUF model quantization, inspect cloud infrastructure via SQL-native queries, audit web accessibility, and execute approval-gated code reviews across various platforms like GitHub, Linear, and Jira.
Who is the target persona for these capabilities?▼
The platform is designed for software engineers, machine learning researchers, and technical project managers who require structured, verifiable, and secure execution of complex development and research tasks.
What are the prerequisites for running these skills?▼
Most skills require a compatible environment supporting Model Context Protocol (MCP) servers, standard shell access for command execution, and specific runtime configurations for libraries like PyTorch, Playwright, or HuggingFace.