subagents-explore-only: read-only instruction for subagents
A rule that uses a SubagentStart hook to tell every subagent to investigate in read-only mode and report findings to the main agent.
Resources that help an AI assistant read, write, review and refactor code.
41,654 results
A rule that uses a SubagentStart hook to tell every subagent to investigate in read-only mode and report findings to the main agent.
A skill covering autonomous agent loops such as ReAct and Plan-Execute, goal decomposition, reflection patterns and production reliability.
A skill for orchestrating distributed training of PyTorch, TensorFlow and HuggingFace models across clusters, with built-in hyperparameter tuning.
A skill for running LLM inference on CPUs, Apple Silicon Macs and consumer AMD or Intel GPUs, without NVIDIA hardware.
A skill for optimizing LLM inference with NVIDIA TensorRT, aimed at production serving on NVIDIA GPUs.
A skill for serving large language models with high throughput using vLLM's PagedAttention and continuous batching.
A skill from davila7 covering LangGraph, a framework for building stateful, multi-actor AI applications with graphs, checkpointed state and human-in-the-loop steps.
A skill with a compact educational GPT implementation of about 300 lines that reproduces GPT-2 (124M) on OpenWebText.
A skill from davila7 for vision-language pre-training that connects frozen image encoders to large language models, released under the MIT license.
A skill that wraps OpenAI's CLIP vision-language model for zero-shot image classification, image-text matching and cross-modal retrieval.
A skill that answers questions about building with OpenAI products and APIs, drawing on official OpenAI documentation with citations.
A skill by davila7 covering image and video processing, object detection, segmentation and visual AI systems.
A skill that guides an AI assistant through building, changing and reviewing user interface code across common frameworks and plain HTML, CSS and Tailwind.
A skill that scans a codebase and produces a quantified technical debt report with a prioritized remediation plan.
A skill that covers Jest testing patterns, factory functions, mocking strategies and the TDD workflow.
A skill for PDF tasks, including text and table extraction, merging, splitting, page rotation, watermarks, form filling, encryption and OCR.
A skill for PDF tasks including text and table extraction, merging, splitting, watermarking, form filling, encryption and OCR.
A skill that uses an append-only audit log of DoorDash CLI calls to answer questions about past agent orders and spending.
A skill for data privacy and regulatory compliance under GDPR, CCPA, HIPAA and other international data protection laws.
An agent for analysing markets, consumer behaviour, competitive landscapes and opportunity size to inform business strategy and market entry decisions.
A skill that frames a senior Quality Manager Responsible Person (QMR) role for HealthTech and MedTech companies, covering quality system governance and regulatory compliance.
A skill that inventories and classifies the git worktrees of the current repository, then removes them after asking the user.
A human-in-the-loop agent for analysing, documenting and planning the modernization of a complete project, with architectural recommendations.
A skill that provides a structured methodology for planning and preparing penetration testing engagements.