Agent Orchestration Skill Leaderboard

Plan engineering work and coordinate parallel agents.

Find and assemble reusable agent workflows.

Customize Copilot workflows for development tasks.

Prototype, test, and develop playable games.

Run engineering, product, marketing, and business workflows.

Preserve task plans and findings across agent sessions.

Design agent context, memory, and evaluation workflows.

Learn and maintain reusable skills from agent sessions.

Coordinate persistent teams of coding agents.

Structure brainstorming and agent development workflows.

Structure agent contracts, payments, and dispute resolution.

Improve projects through measurable autonomous experiments.

Preserve shared agent knowledge in an Obsidian vault.

Add image understanding to text-only agents.

Coordinate software delivery and improve agent workflows.

Explore and critique ideas through parallel reasoning.

Plan projects and coordinate coding agents and tools.

Plan, implement, and verify software changes.

Build a linked knowledge base from agent work and sources.

Develop research arguments and prepare computer science papers.

Coordinate personal tasks across devices with a persistent agent.

Compare independent expert perspectives and synthesize recommendations.

Structure complex reasoning and preserve evidence across agent workflows.

Chat with coding agents through messaging platforms.

Model threats, investigate vulnerabilities, and review security fixes.

Delegate coding tasks across agents and review results.

Design and audit reliable agent architectures.

Organize agent workflows through structured folders and files.

Maintain shared project memory linked to code.

Build customizable Python agents with tools and memory.

Scope, implement, and verify software changes.

Turn product research into plans and stakeholder updates.

Research prospects, qualify leads, and prepare sales outreach.

Improve solutions through competing agents and structured critique.

Delegate code analysis and editing to Codex CLI.

Route agent decisions through TypeSafe's Jev model.

Build agents with LangChain, LangGraph, and Deep Agents.

Build autonomous coding loops with verifiable development environments.

Coordinate coding agents through verified implementation workflows.

Scope, launch, and evaluate Claude Managed Agents.

Coordinate coding-agent handoffs, reviews, and remote validation.

Evaluate decisions through independent advisors and peer review.

Coordinate coding agents and automate browser-based development.

Build games with Godot and Bevy.

Build neuro-symbolic AI agents with Synalinks.

Plan complex tasks and verify agent deliverables.

Run team AI agents with tools and memory.

Plan and deliver customer engineering engagements.

Improve prompts, learning, and multi-agent workflows.

Remember agent mistakes and corrections across coding sessions.