Compose multiple agents as callable tools, spawn dynamic sub-agents at runtime, and wire remote A2A agents into a coordinated pipeline.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Multi-Agent Orchestration
## Agent objective
Produce a builder with agent tools registered so an agent can delegate subtasks to specialized sub-agents and integrate their results.
## When to load this skill
- Building a lead agent that coordinates specialist sub-agents
- Splitting complex tasks across research, writing, coding, or review agents
- Connecting to remote agents via A2A protocol
- Allowing the agent to dynamically spawn sub-agents at runtime based on task needs
## Implementation baseline
```ts
import { ReactiveAgents } from "@reactive-agents/runtime";
// Lead agent with registered sub-agents as tools
const agent = await ReactiveAgents.create()
.withName("lead")
.withProvider("anthropic")
.withReasoning({ defaultStrategy: "plan-execute-reflect", maxIterations: 20 })
.withAgentTool("researcher", {
name: "Research Agent",
description: "Gathers information, reads web pages, and synthesizes findings",
maxIterations: 12,
tools: ["web-search", "http-get", "checkpoint"],
})
.withAgentTool("coder", {
name: "Code Agent",
description: "Writes, tests, and refactors TypeScript code",
maxIterations: 15,
…
The line at the top of this page pins the label v2.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_ybhzsdbjo2gscbze
GET https://markdownregistry.com/api/v1/resolve?ref=tylerjrbuell/reactive-agents-ts/multi-agent-orchestration
GET https://markdownregistry.com/api/v1/blob/83a423873688360119fd61aa67c42938fc3ebe49be26485457f68ed388768bcc
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
tylerjrbuell/reactive-agents-ts · .agents/skills/agent-tdd/SKILL.md · Test-Driven Development for the Reactive Agents codebase. Effect-TS aware TDD with mandatory timeout flags, Effect.flip…
tylerjrbuell/reactive-agents-ts · .agents/skills/architecture-audit/SKILL.md · Use when the architecture may have drifted from documentation, packages have grown complex, dead code or disabled…
tylerjrbuell/reactive-agents-ts · .agents/skills/build-package/SKILL.md · Add a new package to the Reactive Agents monorepo. Covers scaffolding, package.json, tsconfig, layer wiring, and index…
tylerjrbuell/reactive-agents-ts · .agents/skills/codebase-health-sweep/SKILL.md · Use when the codebase may have accumulated bugs, failing tests, type errors, dead code, inefficiencies, or drift from…
tylerjrbuell/reactive-agents-ts · .agents/skills/effect-abstraction-audit/SKILL.md · Use when analyzing the Reactive Agents codebase for architectural improvements, abstraction opportunities…
tylerjrbuell/reactive-agents-ts · .agents/skills/effect-ts-patterns/SKILL.md · Mandatory Effect-TS coding patterns for the Reactive Agents framework. Use when writing any TypeScript code, creating…
tylerjrbuell/reactive-agents-ts · .agents/skills/execute-backlog/SKILL.md · Use when ready to attack the GitHub issue backlog with discipline. Scans open issues, bundles related work into an…
tylerjrbuell/reactive-agents-ts · .agents/skills/harness-improvement-loop/SKILL.md · Harness improvement loop on the H3 frozen suite. SELECT the top harness-attributable failure reason from `rax-diagnose…
tylerjrbuell/reactive-agents-ts · .agents/skills/implement-service/SKILL.md · Create a new Effect-TS service following the mandatory Reactive Agents patterns. Use when adding a service to any…
tylerjrbuell/reactive-agents-ts · .agents/skills/implement-test/SKILL.md · Write tests for a Reactive Agents package or service using Bun test runner and Effect-TS patterns. Use when creating…
tylerjrbuell/reactive-agents-ts · .agents/skills/kernel-debug/SKILL.md · Debug agent reasoning failures using the composable kernel phase map. Maps symptoms to specific files and `repo find`…
jnpiyush/agentx · .github/skills/ai-systems/multi-agent-orchestration/SKILL.md · Design and operate multi-agent systems where several specialized LLM agents collaborate. Use when choosing between…
mehranqadirian/skills · agent-engineering/multi-agent-orchestration/SKILL.md · Designs and runs systems where several AI agents cooperate. Use whenever the user wants to split a task across agents…
int2t05/engineering-skills · skills/04-develop/multi-agent-orchestration/SKILL.md · Execute authorized agent work with explicit ownership and integration checks. Triggers on "parallel agents", "并行 agent".
