Trace, evaluate, and manage LLM application calls, prompts, datasets, and experiments.
Project overview
Langfuse helps teams instrument LLM applications so they can inspect traces, sessions, retrieval steps, prompts, scores, and evaluations in one place. It also supports prompt versioning, datasets, playground experiments, and integrations with common SDKs and frameworks. You can use the hosted service or self-host with Docker Compose or Kubernetes. Self-hosted instances send basic usage telemetry by default, but it excludes raw traces and can be disabled.
Keep long-running agent plans and shared state durable on disk.
Project overview
Store task plans, findings, and progress in Markdown files so coding agents can recover after context loss, session clearing, crashes, or handoffs between workers. Optional lifecycle hooks remind the agent to reread and update the plan, while gated modes prevent it from stopping before explicit completion conditions are met. The skill works across many SKILL.md-compatible coding agents and supports shared state for multi-agent work.
Find and use AI agent skills through a marketplace, hybrid search, and CLI.
Project overview
SkillX combines a web marketplace, command-line client, and semantic-plus-keyword search for discovering agent skills. Users can browse more than 500 entries, compare ratings, reviews, usage statistics, and leaderboards, and save favorites. The current MVP includes authentication, API keys, filters, and pagination, while payments, MCP server mode, and a skill execution sandbox remain future work.
Run resumable, phase-guarded coding workflows and evaluate reusable agent skills.
Project overview
Comet organizes coding work into guarded phases from specification through build, verification, and archive, recording state so an interrupted task can resume without reconstructing its history. It also helps create, compose, distribute, and evaluate reusable skills across many coding agents. The CLI requires Node.js 20 or newer.
Scan agent skills for malicious instructions, unsafe code, and supply-chain risks
Project overview
NVIDIA SkillSpector examines Git repositories, URLs, archives, directories, or individual files before an agent skill is installed. It checks dozens of patterns spanning prompt injection, exfiltration, privilege escalation, dangerous code, dependency vulnerabilities, MCP least privilege, and tool poisoning, then produces terminal, JSON, Markdown, or SARIF reports with risk scores. Scanned skills are never executed. Optional LLM analysis sends file contents to the configured provider, while static-only mode keeps that content local.
Give MCP clients direct access to local terminals, files, edits, and processes.
Project overview
Desktop Commander MCP equips clients such as Claude with local terminal execution, filesystem search and editing, process sessions, long-running commands, and support for formats including Excel, PDF, and DOCX. It can be installed through npx or run in Docker. The server is not itself a security sandbox: allowed-directory settings restrict file tools but do not fully contain terminal commands. Docker isolation and narrowly mounted folders are therefore important when working with untrusted tasks.
Collection or directoryContent creationClaude Code
screen-creative-skills
@GongLingRui·Python
Screen novels, analyze stories, and plan scripts with 31 agent skills.
Project overview
This collection gives compatible AI agents a staged workflow for film and television development, from screening source novels to analyzing plot structure, evaluating adaptation potential, planning vertical dramas, and drafting scripts. Its 31 skills are separated by creative task, while orchestration skills can split long text, coordinate analysis, and combine results. The package covers short-form vertical series, longer dramas, television, and film, but it is an instruction library for writing and evaluation rather than a video production application.
Collection or directoryDesign and creative workClaude Code
Threejs-Awesome-Graphics-Agent-Skills
@scottstts·JavaScript
Give coding agents reusable Three.js graphics systems and implementation examples.
Project overview
Three.js Awesome Graphics Agent Skills packages reusable implementations for cameras, procedural materials and geometry, water, weather, volumetrics, post-processing, and visual validation. Rather than acting as an API cheat sheet, it gives coding agents concrete patterns and example assets for building sophisticated scenes and games. The pack installs through npm into supported hosts including Codex, Claude Code, and Cursor.
Rig and animate Spine characters from separated artwork and references.
Project overview
Spine Animation AI helps Claude turn separated character images and a reference pose into Spine 2D skeletal animation assets. It can position parts with image matching, build skeleton JSON and draw order, generate common motions such as idle or walk, and create interactive HTML5 previews. It can also refine an existing Spine skeleton, but users still need well-prepared artwork and a specific animation goal.
Browse and install design-system skills for AI coding tools.
Project overview
Awesome Design Skills is a curated registry of 67 design skill packages for Claude Code, Cursor, Codex, and other agentic tools. Each package typically pairs an agent-facing SKILL.md with a human-readable DESIGN.md covering brand direction, typography, color, spacing, components, and related visual rules. The TypeUI CLI can pull a chosen skill into a project by slug.
Collection or directoryWriting and editingClaude Code
ResumeSkills
@Paramchoudhary
Optimize resumes and prepare job applications with specialized agent workflows.
Project overview
ResumeSkills covers ATS checks, job matching, truthful resume tailoring, stronger achievement bullets, cover letters, LinkedIn profiles, interview stories, salary negotiation, and offer comparison. It includes workflows for technical, executive, academic, creative, and career-changing candidates. The set is designed for job seekers who want structured help across the application process rather than a single resume rewrite.
Coordinate multiple coding CLIs in one visual parallel workspace.
Project overview
Golutra turns existing coding CLIs such as Claude Code, Codex, Gemini, and OpenCode into a shared multi-agent workspace. Users can run agents in parallel, orchestrate longer workflows, import or export templates, inspect individual logs, inject prompts during execution, and track handoffs in a visual interface. It keeps the underlying CLI tools and project layout intact, making it useful for developers who want centralized oversight without migrating their work to a new agent runtime.
Continue and control local AI coding sessions from other devices.
Project overview
Happier keeps Claude Code, Codex, OpenCode, and other coding-agent sessions running on a computer while making them accessible from a phone, browser, or desktop client. Users can follow, take over, fork, replay, or hand off live sessions without discarding context. Communication is end-to-end encrypted, and the relay can use the hosted service or a self-managed deployment. Remote access requires the target machine and its daemon to remain online; it complements existing agent CLIs rather than replacing their model services.
Collection or directoryProductivity and officeClaude Code
awesome-claude-corporate-skills
@w95·Python
Equip Claude with reusable business playbooks organized by corporate role.
Project overview
Awesome Claude Corporate Skills groups 166 skills across 14 business areas, including leadership, finance, HR, marketing, sales, legal, operations, engineering, product, and analytics. Teams can copy a full role or individual SKILL.md folders into Claude Code, use the repository as a Cowork workspace, or reference the instructions through the Claude API. The collection mixes custom, Anthropic-sourced, and community material, so teams can select and review what fits their standards.
Collection or directorySoftware developmentClaude Code
skills
@NVIDIA·Python
Install NVIDIA-verified agent skills for CUDA-X, AI Blueprints, and platform tools.
Project overview
NVIDIA Agent Skills is an official catalog of portable guidance for NVIDIA software, installable selectively through the skills CLI for clients such as Claude Code, Codex, and Cursor. Entries are mirrored from product repositories and must include signatures, governance cards, evaluation data, and benchmark reports, giving developers a traceable way to add NVIDIA-specific workflows to their agents.
Plan, produce, optimize, and schedule YouTube videos through one pipeline.
Project overview
YouTube Automation Agent links topic planning, script writing, thumbnail and SEO work, media production, publishing, and analytics into a scheduled self-hosted workflow with a local dashboard. Real uploads require YouTube Data API OAuth, assembly requires FFmpeg, and generated text, speech, images, or video depend on configured providers. Without the relevant keys, some stages use templates or simulations; placeholder media is refused by the uploader, and publishing defaults to private.
Collection or directoryResearch and knowledgeClaude Code
Awesome-Gaussian-Skills
@jaccen·HTML
Research, compare, and review 3DGS and NeRF methods with specialized agent skills.
Project overview
Awesome Gaussian Skills combines a searchable knowledge base of hundreds of 3D Gaussian Splatting, NeRF, and graphics methods with a set of focused agent skills. Researchers can use it to read papers, compare techniques across multiple dimensions, review implementations for known bug patterns, plan experiments, migrate NeRF ideas to 3DGS, or bridge CAD, mesh, and Gaussian representations. Each skill is a standalone SKILL.md package for compatible agents.
Find installable Codex skills for development, collaboration, writing, and analysis.
Project overview
Awesome Codex Skills organizes practical skills across development, productivity, communication, data analysis, and utility categories, often with a short use case and installation command. The catalog mixes skills stored in the repository with links to external projects, so it is useful for discovery but each item’s permissions, dependencies, and quality still need separate review.