Search Results for "collection"

Agent Skills for Context Engineering is a curated collection of reusable “agent skills” focused on helping AI agents perform better on long-horizon, multi-step work by managing context deliberately. Rather than being a single application, it packages practical guidance into skill modules that agents can load to improve planning, retrieval, memory usage, and overall reliability in real workflows.

Scientific Agent Skills is an open-source collection of ready-to-use agent skills designed to turn AI coding assistants into stronger research, science, engineering, and analysis partners. It supports any AI agent compatible with the Agent Skills standard, including tools such as Cursor, Claude Code, Codex, and Gemini CLI. The repository includes 135 skills across scientific domains such as genomics, cheminformatics, clinical research, medical imaging, machine learning, physics, materials science, geospatial analysis, and scientific writing. ...

Agent Lightning is an open-source framework developed by Microsoft to train and optimize AI agents using techniques like reinforcement learning (RL), supervised fine-tuning, and automatic prompt optimization, with minimal or zero changes to existing agent code. It’s designed to be compatible with a wide range of agent architectures and frameworks — from LangChain and OpenAI Agent SDKs to AutoGen and custom Python agents — making it broadly applicable across different agent tooling...

Qwen-MM-Plugins is a collection of native multimodal capabilities that can extend Qwen-powered agent harnesses beyond text-only workflows. Each capability combines an instructional skill with an optional MCP server that exposes specialized tools. The core package can inspect images, videos, documents, and 3D models while providing OCR, grounding, segmentation, transcription, and visual question answering.

ADR, short for Agentic AI Detection and Response, is an enterprise security system for monitoring and evaluating AI agents. It captures agent intent, tool activity, and execution traces from coding assistants, internal automations, and customer-facing agents. A normalized sensor layer provides observability across multiple agent tools and operating systems. ADR-Bench supplies more than 300 realistic tasks, 133 MCP servers, and coverage of 17 documented agent attack techniques. Its two-tier...

Claude for Financial Services is an open-source collection of AI agents, plugins, and workflow templates designed to transform Claude into a specialized assistant for financial services professionals. The project targets domains such as investment banking, equity research, private equity, and wealth management by providing reusable prompts, structured workflows, and domain-specific analytical skills.

Khazix Skills project is an automation framework designed to transform GitHub repositories into structured, reusable AI agent skills. It acts as a pipeline that analyzes a repository’s metadata, extracts relevant information such as README content and commit hashes, and converts it into a standardized skill format that can be integrated into agent ecosystems. The system emphasizes lifecycle management by embedding versioning, traceability, and metadata directly into generated skill files,...

Live Agent Studio is a curated repository of open-source AI agents associated with the oTTomator Live Agent Studio platform, showcasing a variety of agent implementations that illustrate how autonomous and semi-autonomous tools can be constructed using modern AI frameworks. Each agent in the collection is designed for a specific use case — such as content summarization, task automation, travel planning, or RAG workflows — and is provided with the code or configuration needed to explore and extend it on your own, making the repository both a learning resource and a practical starting point for real projects. The repository is community focused, with sample agents like tweet generators, smart selectors, research assistants, and multi-tool workflows that show how agents can integrate with tools like n8n or custom Python code. ...

...To achieve this, OmniParser curates an interactable icon detection dataset containing 67,000 unique screenshot images labeled with bounding boxes of interactable icons derived from DOM trees. Additionally, a collection of 7,000 icon-description pairs is used to fine-tune a caption model that extracts the functional semantics of detected elements. Evaluations on benchmarks such as SeeClick, Mind2Web, and AITW demonstrate that OmniParser outperforms GPT-4V baselines, even when using only screenshot inputs without additional information.