Find your next open-source project

Search by name, purpose, or keyword — or paste a GitHub repository URL
Back

ruvnet/ruflo

Provides agent skills for running coordinated multi-agent swarms, shared memory, and automated workflows in ruflo.

  • 182 skills
  • 79k stars

The 5 most installed

Ordered by installs on skills.sh
  • ruflo4.2k installs

    Installs and runs Ruflo's multi-agent swarms, tools, plugins and health checks in your project.

  • swarm-orchestration2.1k installs

    Coordinates specialized agents to split and execute complex multi-file changes across a codebase.

  • Runs an agent that designs system architecture.

  • agent-workflow1.4k installs

    Runs a defined multi-step workflow across agents.

  • Manages GitHub issues, project boards and sprint planning with agent coordination.

The other 177 skills
  • security-audit1.4k installs

    Scans code for vulnerabilities like injection, path traversal and known CVEs, and enforces safe patterns.

  • Writes specifications for mobile apps built with React Native.

  • agent-swarm1.3k installs

    Launches a coordinated swarm of agents that work on a task together.

  • Builds trading predictions from market data using agent workflows.

  • github-code-review1.3k installs

    Runs AI-coordinated code reviews on GitHub pull requests.

  • Builds GitHub Actions CI/CD pipelines with agent-coordinated automation.

  • workflow-automation1.3k installs

    Creates, runs, and manages reusable multi-step workflows with agent coordination.

  • Runs an agent that handles agentic payment flows.

  • Runs an agent that analyzes code quality and reports issues.

  • agent-architecture1.2k installs

    Runs an agent that works on software architecture tasks.

  • agent-code-analyzer1.2k installs

    Runs an agent that analyzes source code.

  • agent-data-ml-model1.2k installs

    Builds and runs machine learning models for data analysis tasks.

  • Writes OpenAPI specifications documenting an API's endpoints and schemas.

  • Sets up and runs CI/CD pipelines on GitHub for building and deploying code.

  • Orchestrates multi-step tasks by assigning subtasks to the right agents.

  • agent-payments1.2k installs

    Handles payment flows and billing logic in an application.

  • Watches system and agent performance and reports slowdowns or bottlenecks.

  • Finds performance bottlenecks and suggests or applies speed improvements.

  • agent-planner1.2k installs

    Breaks a task into an ordered plan with clear steps before work starts.

  • Designs repository structure and layout for a codebase.

  • agent-researcher1.2k installs

    Researches a topic and gathers findings to support a task.

  • Manages shared memory so a swarm of agents can share context.

  • agent-tester1.2k installs

    Writes and runs tests for your code and reports failures.

  • Automates repeated multi-step agent workflows.

  • agentic-jujutsu1.2k installs

    Manages version control for AI agents with self-learning and multi-agent coordination.

  • embeddings1.2k installs

    Creates fast vector embeddings for semantic search, pattern matching and knowledge retrieval.

  • flow-nexus-swarm1.2k installs

    Deploys cloud AI swarms and event-driven workflows on the Flow Nexus platform.

  • github-automation1.2k installs

    Automates GitHub pull requests, issues, code review and release workflows.

  • github-multi-repo1.2k installs

    Coordinates and synchronizes changes across multiple GitHub repositories.

  • Orchestrates GitHub releases with versioning, testing, deployment and rollback.

  • memory-management1.2k installs

    Stores and searches past solutions in a fast vector memory so agents can reuse what worked.

  • Pair Programming1.2k installs

    Pairs with you as driver or navigator, switching roles while reviewing code and running tests.

  • pair-programming1.2k installs

    Pairs with you as driver or navigator, switching roles while reviewing code and running tests.

  • Analyzes swarm performance, finds bottlenecks and suggests optimizations.

  • Skill Builder1.2k installs

    Creates new Claude Code skills with proper frontmatter, structure and directory layout.

  • skill-builder1.2k installs

    Creates new Claude Code skills with proper structure, metadata, and directory layout from your requirements.

  • sparc-methodology1.2k installs

    Guides a feature through specification, pseudocode, architecture, refinement, and completion before coding.

  • stream-chain1.2k installs

    Chains multiple agents and data transformations into sequential streaming pipelines.

  • swarm-advanced1.2k installs

    Runs advanced multi-agent swarm patterns for research, development, testing, and distributed workflows.

  • Runs an adaptive coordinator agent that adjusts how work is spread across agents.

  • agent-agent1.1k installs

    Invokes a general-purpose agent for tasks that do not fit a specialized role.

  • agent-app-store1.1k installs

    Runs an agent focused on app store publishing tasks.

  • Runs an agent that handles authentication work.

  • Runs a smart agent that automates routine tasks.

  • Runs an agent that generates starter project templates.

  • Runs an agent that executes benchmark suites and reports results.

  • Runs a coordinator agent that keeps agreement when some agents misbehave.

  • agent-challenges1.1k installs

    Runs an agent that generates and manages challenge tasks.

  • Runs an agent that plans coding work toward a stated goal.

  • Runs a swarm of agents that review code together.

  • agent-coder1.1k installs

    Runs an agent that writes code.

  • Runs a coordinator agent that combines input from many agents.

  • Runs a coordinator agent that drives agents to reach consensus.

  • Sets up a new swarm of coordinated agents with an initial topology and roles.

  • Keeps shared state consistent across agents using conflict-free replicated data types.

  • Builds backend API endpoints and services for an application.

  • agent-github-modes1.1k installs

    Switches between GitHub working modes such as reviewing, triaging and releasing.

  • Manages GitHub pull requests, including reviews, labels and merge steps.

  • agent-goal-planner1.1k installs

    Breaks a stated goal into concrete steps and an execution plan.

  • Coordinates agents by spreading updates through a gossip-style message protocol.

  • Runs agents in a hierarchy where leaders assign and track work for subordinates.

  • Implements code following the SPARC methodology's specification and pseudocode steps.

  • agent-issue-tracker1.1k installs

    Tracks issues through their lifecycle and keeps their status up to date.

  • agent-load-balancer1.1k installs

    Distributes tasks across agents to balance workload and avoid bottlenecks.

  • Optimizes matrix computations used in numerical and machine learning workloads.

  • Coordinates shared memory so multiple agents can store and retrieve context.

  • Coordinates agents in a mesh topology where peers communicate directly.

  • Drafts a step-by-step plan for migrating a system or codebase.

  • Coordinates a swarm of agents working across several repositories at once.

  • Designs and trains neural network models for a given task.

  • Ranks code or graph nodes by importance using PageRank-style analysis.

  • Analyzes system or code performance and points out slow spots.

  • Runs benchmarks and reports performance numbers for comparison.

  • agent-pr-manager1.1k installs

    Manages pull requests, from opening them to shepherding review and merge.

  • Checks that changes are safe and ready before they reach production.

  • Keeps project board cards in sync with issues and pull requests.

  • agent-pseudocode1.1k installs

    Writes pseudocode to sketch logic before real code is written.

  • Acts as a lead coordinator that assigns work across a swarm of agents.

  • Runs quorum-based decisions so multiple agents agree before acting.

  • agent-raft-manager1.1k installs

    Coordinates agents using the Raft consensus protocol for reliable agreement.

  • agent-refinement1.1k installs

    Refines rough requirements or drafts into clearer, more precise versions.

  • Handles release steps such as versioning, changelogs and publishing.

  • agent-release-swarm1.1k installs

    Coordinates several agents to prepare and ship a software release.

  • Assigns agents and compute resources to tasks based on need.

  • agent-reviewer1.1k installs

    Reviews code changes and points out problems or improvements.

  • agent-safla-neural1.1k installs

    Runs neural learning routines that help agents adapt from experience.

  • agent-sandbox1.1k installs

    Runs agent work in an isolated sandbox to keep things safe.

  • Explores an unfamiliar codebase or system and reports what it finds.

  • Checks code and configuration for security risks and weak spots.

  • Tunes agent learning settings to improve results over time.

  • Coordinates agents through the SPARC planning and development method.

  • agent-specification1.1k installs

    Turns an idea into a written specification with requirements and scope.

  • agent-swarm-issue1.1k installs

    Handles issues by splitting the work across a swarm of agents.

  • agent-swarm-pr1.1k installs

    Coordinates a swarm of agents to prepare and handle pull requests.

  • Keeps multiple agents in sync while they work on shared tasks.

  • Runs a swarm of agents that write tests first, London style, then code.

  • Runs long test suites and reports back when they finish.

  • Chooses the best way to arrange agents for a given task.

  • agent-user-tools1.1k installs

    Provides helper tools an agent can call to assist users.

  • Plans how to integrate the v3 system with existing components.

  • Designs and manages the memory layer for the v3 system.

  • Finds and fixes performance problems in the v3 system.

  • Acts as the lead agent that directs others in the v3 system.

  • Reviews and designs security for the v3 system.

  • Runs a worker agent that carries out assigned subtasks.

  • claims1.1k installs

    Grants, revokes and verifies permissions so agents can coordinate securely.

  • flow-nexus-neural1.1k installs

    Trains and deploys neural networks in distributed sandboxes with Flow Nexus.

  • flow-nexus-platform1.1k installs

    Manages Flow Nexus accounts, sandboxes, app deployment, payments and challenges.

  • hive-mind1.1k installs

    Runs fault-tolerant multi-agent consensus with a queen-led hierarchical swarm.

  • hive-mind-advanced1.1k installs

    Coordinates a queen-led team of agents that reach decisions by consensus and keep a shared memory.

  • Hooks Automation1.1k installs

    Runs automatic hooks around tasks to format code, sync memory and learn from each session.

  • hooks-automation1.1k installs

    Runs automatic hooks around tasks to format code, sync memory and learn from each session.

  • neural-training1.1k installs

    Trains and consolidates neural patterns for adaptive routing and knowledge transfer between tasks.

  • Adds adaptive learning so agents recognize patterns, refine strategies and improve over time.

  • Adds adaptive learning so agents recognize patterns, refine strategies and improve over time.

  • Modernizes the claude-flow v3 command line with interactive prompts, command splitting, and hook integration.

  • Implements claude-flow v3 core modules using domain-driven design, dependency injection, and TypeScript tests.

  • V3 DDD Architecture1.1k installs

    Builds claude-flow v3 architecture around domain-driven design, bounded contexts, and a microkernel pattern.

  • V3 MCP Optimization1.1k installs

    Optimizes the claude-flow v3 MCP server with connection pooling, load balancing, and faster tool responses.

  • Merges claude-flow v3 memory systems into AgentDB with fast HNSW vector indexing.

  • Benchmarks and tunes claude-flow v3 for faster attention, quicker search, and lower memory use.

  • Reworks claude-flow v3 security to fix known vulnerabilities and apply secure-by-default patterns.

  • Coordinates 15 specialized agents in parallel to implement claude-flow v3 across security, core, and integration.

  • Modernizes the claude-flow v3 command line with interactive prompts, command splitting, and hook integration.

  • Implements claude-flow v3 core modules using domain-driven design, dependency injection, and TypeScript tests.

  • v3-ddd-architecture1.1k installs

    Builds claude-flow v3 architecture around domain-driven design, bounded contexts, and a microkernel pattern.

  • v3-mcp-optimization1.1k installs

    Optimizes the claude-flow v3 MCP server with connection pooling, load balancing, and faster tool responses.

  • Merges claude-flow v3 memory systems into AgentDB with fast HNSW vector indexing.

  • Benchmarks and tunes claude-flow v3 for faster attention, quicker search, and lower memory use.

  • Reworks claude-flow v3 security to fix known vulnerabilities and apply secure-by-default patterns.

  • Coordinates 15 specialized agents in parallel to implement claude-flow v3 across security, core, and integration.

  • worker-benchmarks1.1k installs

    Runs benchmarks on the worker system and reports performance analysis.

  • worker-integration1.1k installs

    Connects workers with agents for task dispatch and performance tracking.

  • agentdb-query746 installs

    Queries AgentDB memory for semantic routing, recall, causal graphs and stored patterns.

  • adr-create743 installs

    Creates a new numbered architecture decision record and registers it in AgentDB.

  • wasm-agent741 installs

    Creates and manages sandboxed WASM agents that run code in isolation.

  • adr-index738 installs

    Builds or rebuilds the architecture decision record index and dependency graph in one script run.

  • federation-status731 installs

    Summarizes the health, channels, claims and active agents of the public federation gateway.

  • federation-audit730 installs

    Reviews signed federation messages, channel activity and work claims for inconsistencies.

  • federation-init730 installs

    Walks you through connecting to RuFlo Federation and its available channels.

  • adr-verify689 installs

    Reads back decision record data and reports dangling references, supersede cycles and unknown statuses.

  • browser674 installs

    Automates web browsing and captures AI-friendly page snapshots for agents.

  • managed-agent597 installs

    Runs a Claude Managed Agent in the cloud with its own container, files and tools.

  • nested-subagents317 installs

    Spawns sub-agents that can spawn their own sub-agents, up to five levels deep.

  • adr-reindex64 installs

    Rebuilds the decision record index from scratch to clear stale entries after deletions.

  • agntcy-status42 installs

    Reports whether AGNTCY/SLIM integration is installed and configured this session.

    Runs only when you type /agntcy-status

  • open-federation9 installs

    Coordinates with an open swarm federation online, checking who is active and assigning shared work.

  • Creates, assigns and updates tasks so each piece of team work has one owner.

  • launch-ai-team2 installs

    Plans an AI team for a goal, assigning roles, tasks and a budget without executing any work.

  • Pauses, resumes or completes an AI team or run while keeping its recorded evidence.

  • monitor-ai-team2 installs

    Reports an AI team's progress, blockers, task owners and budget use without changing anything.

  • Saves approved decisions and facts for one AI team and recalls them later with their source.

  • Checks an AI team's work against its acceptance criteria, separating verified evidence from claims.

  • Builds self-learning agents with AgentDB's reinforcement learning algorithms.

  • Trains AI agents to learn from experience using nine reinforcement learning algorithms.

  • Checks code changes against accepted architecture decision records and flags violations.

  • Spawns and manages 60+ agent types and coordinates multi-agent tasks and pools.

  • Sets up advanced AgentDB features like sync, hybrid search and custom metrics.

  • Adds persistent session and long-term memory to agents using AgentDB.

  • Speeds up AgentDB with quantization, indexing, caching and batch operations.

  • Sets up advanced AgentDB features like sync, hybrid search and custom metrics.

  • Gives agents persistent memory for sessions, long-term storage and context management.

  • Speeds up AgentDB memory and search with quantization, indexing, caching and batching.

  • Coordinates a RuFlo agent swarm by reading updates, checking claims and planning next steps.

  • Finds and runs remote workflows with policy checks, worker inspection and acceptance gates.

  • Prepares and posts a short federation update when you ask to announce status or results.

  • Tracks agent trajectories and distills them into reusable patterns using a fast vector database.

  • Tracks agent trajectories and distills them into reusable patterns using a fast vector database.

  • Inspects swarm channels and work claims to find blockers, collisions, and the best next task.

  • Integrates claude-flow v3 with agentic-flow as an extension, removing duplicated code.

  • Integrates claude-flow v3 with agentic-flow as an extension, removing duplicated code.

  • Scores code output for correctness and quality, and rolls back changes that fall below a set accuracy threshold.

  • Checks a code change's risk and complexity using ruflo's analysis commands, and adds CI regression guards.