Agent Orchestration Skill Leaderboard

A large-scale collection of agentic skills and templates for building specialized AI automation capabilities.

Provides file-based persistence for AI agents using task_plan.md, findings.md, and progress.md to track multi-step projects across session resets.

A 73-skill game development pipeline covering brainstorming, design, architecture, prototyping, implementation, testing, QA, localization, security, release, and live ops for indie teams and solo developers.

A collection of agent skills focused on context engineering, harness design, and multi-agent architecture for production systems.

A set of agent orchestration skills covering self-scheduling, multi-agent delegation loops, and direct browser automation via CDP.

Parallel divergent ideation for coding agents. Spawns isolated branches under cognitive frames, scores, clusters, prunes traps, and deepens top survivors for open-ended design and architecture problems.

Fifteen skills that analyze a website and produce audits, copy, ads, emails, social calendars, landing page CRO, competitor intelligence, launch plans, proposals, reports, SEO audits, brand guidelines, and PDF reports from a single command.

Architectural patterns for provider-neutral agent loops, tool design, and production-safe autonomous harnesses.

Implements a think-act-prove orchestration method for AI agents featuring domain research and adversarial verification of outputs.

14 legal skills for contract review, risk analysis, comparison, plain English translation, counter-proposals, document generation, and compliance auditing with PDF reporting.

Runs high-stakes decisions through five independent AI advisors with peer review and synthesis for a clear verdict and actionable next step.

Fifty-two skills for empirical economics: data analysis, DiD event studies, Monte Carlo simulations, LaTeX/Quarto slides, manuscript review, replication packages, grant proposals, and preregistration.

Over thirty curated skills for marketing lead research, competitive ad analysis, content writing, generative art, frontend development, Notion integration, document processing, and operational workflow automation.

A comprehensive set of prompting skills for product managers to generate PRDs, accessibility audits, and professional feedback reports.

Agent orchestration skills for provisioning isolated execution environments, auditing agent context, and setting up reproducible development harnesses.

Two-skill workflow for planning and writing philosophy/interdisciplinary papers: platform analysis, literature search, gap identification, outline optimization, then systematic writing with quality gates.

A three-phase plan hardening tool that uses Claude for requirements interrogation and OpenAI Codex for adversarial stress-testing of implementation plans.

Translates PDF, DOCX, and EPUB books into multiple languages using a parallel sub-agent workflow and Markdown chunking.

A collection of 14 sales skills covering prospect research, lead qualification, decision maker identification, cold outreach sequences, follow-ups, meeting prep, proposals, competitor intelligence, and pipeline reporting.

Provides delegation skills that allow an orchestrator agent to assign coding tasks to specialized CLI agents including Aider, Claude Code, and Cline.

A knowledge management system featuring parallel research decomposition, intelligent thought capture, and a V-model verification lifecycle for AI agents.

Covers idiomatic Synalinks development: data models, program APIs, modules, control flow, agents, RAG, training, rewards, and providers.

LLM-first SEO audits covering technical SEO, content quality, schema, performance, AI search, and GitHub repos with scored reports and action plans.

An agent orchestration framework that enforces staged execution, sub-agent delegation, and verification to prevent fabrication in complex tasks.

A collection of security review skills for AI agents to autonomously find, reproduce, and patch software vulnerabilities.

Claude Code skills for structured planning, collaboration, debugging, testing, code review, problem-solving, and custom skill authoring.

A set of orchestration tools that convert run history, failure traces, and refined workflows into reusable AI skill packages.

Provides language-agnostic coding principles, backend technical decision criteria, and anti-pattern detection for code reviews and feature implementation.

Automates academic paper drafting by converting agent logs into LaTeX manuscripts and refining them via simulated peer review cycles.

An eleven-chapter reference skill for LangGraph 1.0 covering ReAct agents, StateGraph, middleware, human-in-the-loop, memory, supervisor patterns, RAG with vector/keyword/hybrid retrieval, and parallel execution.

A comprehensive ASO toolkit for keyword research, metadata optimization, A/B testing, review management, and launch strategy on Apple App Store and Google Play Store.

Queries ChatGPT and Gemini APIs for implementation perspectives, synthesizes their responses, and presents a plan informed by multiple LLMs.

Implements knowledge graph construction—including ontology design and entity fusion—and multi-agent task orchestration using graph-based workflows.

Coordinates Claude, Codex, and Gemini as an agent team for multi-perspective code and content reviews, with separate roles for bugs, architecture, readability, and logic.

A tool for orchestrating multi-agent GPT workflows using a DSL to manage parallel pipelines and phases for codebase audits.

Audits, distills, and maintains OpenClaw workspace files—AGENTS.md, SOUL.md, TOOLS.md, MEMORY.md, and checklists—for token efficiency and cross-file consistency.

An engagement fieldbook for Forward Deployed Engineers to track client memory, stakeholder trust, and project deliverables.

Reconstructs and continues prior AI coding-agent sessions from any major platform, with provenance checks, task extraction, and workspace validation.

A source-cited playbook for building AI agents on AWS, covering Strands, Bedrock, AgentCore, Terraform, IAM, guardrails, and production deployment.

A review tool that interrogates implementation plans against documentation and source evidence to identify unknown unknowns.

A multi-role agent orchestrator using digraph-based pipelines to build, analyze, and review code with mandatory independent evaluation.

Provides a file-based memory system for agents to record progress and restore project context across multiple sessions.

Uses three isolated subagents—Hunter, Skeptic, Referee—to find, challenge, and verify code bugs, delivering a severity-sorted report of confirmed findings.

Establishes a Codex-native self-improving memory loop with five core files, AGENTS.md integration, and optional nightly refinement automation.

A toolset for creating, updating, and querying Linear issues, projects, and team workflows.

An orchestration protocol that uses plan gates, adversarial verification, and strict execution cards to improve the coding quality of non-frontier AI models.

An agent orchestration tool that offloads repetitive tasks to cheaper models based on live benchmarks and token limits.

Delegates coding, review, analysis, and research to Gemini sub-agents in background with fan-out, isolation, and verification workflows.

A structured workflow for planning, subagent use, lessons tracking, and verification that enforces disciplined task execution with quality gates.