Agent Workforce
Official@agentworkforce
Building the future of AI agent workforce orchestration for software development
Agent Skills by Agent Workforce
Showing 27 vetted skills indexed across 6 GitHub repositories.
agent-relay
Coordinate multiple Codex agents through Relaycast messaging with ACK and DONE signaling.
creating-agent-skills-skill
Create SKILL.md agent skill packages following the agentskills.io specification.
using-agent-relay
Coordinates registered agents over Agent Relay MCP tools for messaging, channels, and worker management.
review-animations
Reviews animation and motion code against craft standards for easing, duration, performance, and accessibility.
emil-design-eng
Reviews and improves UI animations, CSS transitions, and component interactions using design engineering principles.
github-oauth-nango-integration
Implements dual-connection GitHub OAuth and GitHub App authentication with Nango webhooks.
deploying-to-staging-environment
Coordinates multi-repo staging deployments across relay, relay-dashboard, and relay-cloud using git worktrees and GitHub Actions.
browser-testing-with-screenshots
Automates Chrome browser interactions and screenshot capture for visual UI verification.
debugging-websocket-issues
Diagnose WebSocket frame errors caused by conflicting WebSocketServer instances and compression misconfiguration.
frontend-design
Generate production-grade frontend interfaces with distinctive typography, color, motion, and layout.
review-fix-signoff-loop
Repair and re-validate workflow implementations until dual signoff agents agree.
relay-80-100-workflow
Validate agent-relay workflows end-to-end with PGlite and mock sandboxes.
using-trajectories-sdk
Capture and persist agent work trajectories with TrajectoryClient and TrajectoryBuilder.
running-headless-orchestrator
Bootstrap relay infrastructure and coordinate autonomous worker teams.
optimizing-smithery-score
Optimize Smithery MCP quality scoring for published MCP servers.
choosing-swarm-patterns
Select among 10 orchestration patterns for coordinating AI agents.
writing-agent-relay-workflows
Design multi-agent workflows with the Relay broker-sdk across TypeScript, Python, or YAML.
relayfile-workspace
Coordinate multi-agent file operations on a Relayfile shared filesystem.
gitclawlab
Manage Git repositories and deploy Dockerfile-based applications via REST API.
frontend-design
Create production-grade React landing pages with accessible markup and responsive layouts.
prpm-json-best-practices
Structure prpm.json manifests for publishing packages and collections.
implementing-command-palettes
Implement Cmd+K command palettes in React with keyboard navigation and filtering.
creating-claude-agents
Create Claude Code agents with frontmatter, persona, tools, and schema validation.
using-agent-relay
Coordinate multiple AI agents with agent-relay and a file-based inbox.
Frequently Asked Questions About Agent Workforce
FAQPage SchemaWhat specific development tasks are enabled by these capabilities?βΌ
These capabilities enable the construction of production-grade React interfaces, the implementation of keyboard-navigable command palettes, and the orchestration of multi-agent teams. You can manage complex repository deployments, validate code through dual signoff loops, and persist work trajectories using structured JSON manifests and file-based coordination.
Which personas benefit most from these technical skills?βΌ
Software engineers, technical leads, and systems architects focused on building autonomous development environments benefit most. These skills are designed for developers managing multi-agent systems, those configuring Claude Code environments, and teams requiring standardized repository management and high-quality MCP server publishing.
What are the prerequisites for implementing these orchestration patterns?βΌ
Implementation requires familiarity with TypeScript, YAML, and JSON I/O conventions. Users must be able to structure prpm.json manifests, configure Claude Code frontmatter, and manage Dockerfile-based deployments. A foundational understanding of file-based inbox coordination and PGlite sandbox testing is necessary for end-to-end validation.