an open-source alternative to the dots, bots and muses of the world, run an office of claude code/codex like agents on your laptop, sandboxes or anywhere, uses your existing subscriptions
HTML agent-harnessagent-orchestrationagentsai-agentsautonomous-agentsclaude-codecodexcoding-agentsdesktop-appdeveloper-toolselectrongemini-cliharnessharness-engineeringlocal-firstmemorymulti-agentopencodeorchestrationtypescript
blog: multi-agent-orchestration-patterns (78 → 79)
Chaitanya Giri805f44cc8f07
diff --git a/blog/src/posts/multi-agent-orchestration-patterns.md b/blog/src/posts/multi-agent-orchestration-patterns.mdnew file mode 100644index 00000000..c216a103--- /dev/null+++ b/blog/src/posts/multi-agent-orchestration-patterns.md@@ -0,0 +1,137 @@+---+title: "Multi-Agent Orchestration Patterns: Which Topology, and When"+description: "The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins."+date: 2026-06-05+category: orchestration+categoryLabel: Orchestration+type: Technical+primaryKeyword: "multi-agent orchestration patterns"+secondaryKeywords: ["agent orchestration topologies", "orchestrator worker pattern", "supervisor agent pattern", "when to use multi-agent"]+tags: ["Orchestration", "Multi-Agent", "Concepts", "Architecture"]+author:+ name: Chaitanya Giri+ initials: CG+faq:+ - q: "What are the main multi-agent orchestration patterns?"+ a: "Six recur: orchestrator-worker (a.k.a. supervisor — a central agent decomposes and delegates), hierarchical (supervisors of supervisors), sequential pipeline (fixed linear stages), parallel fan-out (scatter-gather over independent tasks), debate/maker-checker (one agent produces, another verifies), and blackboard (agents coordinate through shared state). Most production systems lean on orchestrator-worker."+ - q: "Which orchestration pattern should I use?"+ a: "Match it to your dependency structure: linear dependencies → pipeline; independent tasks → fan-out; accuracy-critical work → maker-checker; known sub-tasks with one accountable owner → orchestrator-worker (the 2026 default). Start with the simplest pattern that fits — most teams over-architect and add agents they don't need."+ - q: "Is a multi-agent system always better than a single agent?"+ a: "No. In benchmarks, a single agent matched or beat multi-agent systems on a majority of tasks when given the same tools and context, and a large share of multi-agent pilots fail in production — usually from the wrong pattern. More agents add coordination cost; they pay off only when the task genuinely decomposes."+---++<div class="callout tldr"><span class="ic">TL;DR</span><p>Multi-agent systems come in a few named+<strong>topologies</strong> — orchestrator-worker (supervisor), pipeline, fan-out, debate/maker-checker,+swarm, and blackboard. <strong>Orchestrator-worker is the 2026 default.</strong> But the most important+fact is the anti-hype one: a <strong>single agent often matches or beats</strong> a multi-agent system on+the same task, and many multi-agent pilots fail — usually from picking the wrong pattern. Rule: start with+the simplest topology that fits, add agents only when the problem genuinely decomposes, and make whatever+you pick <strong>survivable</strong>.</p></div>++Once you've decided you need more than one agent, the next question is *how they're wired together* —+the topology. There's a small, well-worn set of patterns, each suited to a different shape of problem.+This is the catalog, when to reach for each, the honest caveats, and how the patterns show up in a real+[multi-agent hive](/#how).++## The pattern catalog++- **Orchestrator-worker (supervisor).** A central orchestrator classifies the task, decomposes it,+ dispatches sub-tasks to specialized workers, and merges the results. One clear accountability point, and+ workers can run cheaper task-specific models — say, a research orchestrator that dispatches lookups to+ small, fast workers and reserves a strong model for the final synthesis. **Use when** you know the+ sub-tasks at design time and want one place that owns the outcome. This is the 2026 production workhorse.+- **Hierarchical (supervisors of supervisors).** Tiered orchestrator-worker: higher tiers plan and+ coordinate, lower tiers execute. **Use when** the work is too big for one orchestrator's context — teams+ of teams.+- **Sequential pipeline.** Fixed linear stages, each consuming the previous output. **Use when** the order+ never changes and steps strictly depend on each other (extract → transform → validate).+- **Parallel fan-out (scatter-gather).** Dispatch N independent sub-tasks at once, then gather. **Use+ when** you have several tasks with no dependencies between them — the classic latency win.+- **Debate / maker-checker.** One agent produces, another critiques or verifies; iterate or vote. **Use+ when** accuracy matters more than speed — review loops, adversarial checks.+- **Swarm / network (peer).** Dynamic peer agents hand off to each other with no central conductor. **Use+ when** the path is emergent and data-dependent — and budget for it being the hardest to debug and bound.+- **Blackboard.** Agents read and write a shared workspace rather than messaging directly; the shared+ state coordinates them. **Use when** many agents contribute to one evolving artifact.++## A decision framework++You can get most of the way with three rules:++1. **Default to supervisor / orchestrator-worker.** It's legible, accountable, and debuggable — which is+ why it dominates production.+2. **Match the topology to the dependency structure.** Linear deps → pipeline. No deps → fan-out.+ Accuracy-critical → maker-checker. Emergent path → swarm (with guardrails).+3. **Start simple.** Most teams over-architect. Reach for more agents only when a single agent genuinely+ can't hold the task.++## Patterns compose++Real systems rarely use one pattern in isolation — they nest. A supervisor's workers might each run a+maker-checker loop internally; a sequential pipeline might fan out within a single stage; a hierarchical+team might keep a blackboard at each tier to pool findings. The patterns are a *vocabulary*, not a menu you+pick one item from. The discipline is the same at every level of nesting: at each composition point, ask+what the dependency structure of *that* sub-problem is, and choose the topology that fits it.++A common, effective shape is a **supervisor on the outside** — one accountable owner with clear+decomposition — wrapping **fan-out and maker-checker on the inside**: run the independent sub-tasks in+parallel, and route the risky or accuracy-critical ones through a verifier. Compose deliberately, though:+every boundary you add is another place coordination can fail, and another handoff to debug.++## The honest part: multi-agent isn't free++Two findings are worth leading with, so this stays grounded and not hype:++- **A single agent often wins.** In benchmarking, a single agent matched or outperformed multi-agent+ systems on a majority of tasks when given the same tools and context. More agents add coordination+ overhead, latency, and cost — the payoff only appears when the task truly decomposes.+- **Many pilots fail.** A large share of multi-agent projects don't survive their first months in+ production, commonly because the team picked the wrong pattern, or the right pattern without+ understanding its failure mode — a swarm with no hop cap loops; a fan-out with hidden cross-task+ dependencies corrupts its merge.++The corollary: the value isn't "more agents," it's **the right topology plus the resilience to survive+its failure mode** — hop caps, idempotent handling, isolation. (We wrote up those mechanics in+[recovering from agent failures](/blog/recovering-from-agent-failures/).)++## The 2026 framework landscape++If you're building this on a framework, the control-vs-simplicity tension hasn't converged — they differ+mostly in orchestration model and ecosystem:++- **LangGraph** — a directed graph with conditional edges; the most control, the most boilerplate. The+ default for stateful, auditable, regulated workflows.+- **CrewAI** — role-based "crews" with process types; the fastest path to a working multi-agent team.+- **AutoGen / AG2** — conversational group chat between agents.+- **OpenAI Agents SDK** — explicit handoffs between agents, with a built-in execution sandbox.+- **Claude Agent SDK / Strands** — simplicity over fine-grained orchestration.++The 2026 trend is convergence on common abstractions (graphs, roles, handoffs) with differentiation in+ecosystem depth and ops tooling. Pick for the orchestration model you need, not the logo.++## How the patterns map onto a hive++A concrete system makes the patterns less abstract. Munder Difflin runs several at once:++- **Orchestrator-worker / supervisor** — the [god orchestrator](/blog/how-the-god-orchestrator-works/)+ classifies intent, decomposes it, dispatches to role-specialized agents, and owns integration. That's+ the 2026 default, made local and watchable.+- **Blackboard** — the shared board and the append-only event log are a blackboard the team reads and+ writes.+- **Fan-out** — a broadcast message is scatter-gather over the active roster.+- **Maker-checker** — human-in-the-loop approval and adversarial review act as a consensus gate on risky+ output.++It's the same lesson throughout: orchestration is how agents [coordinate without+colliding](/blog/coordinating-ai-coding-agents/), and the pattern you choose is a big part of whether they+do.++## Pick the topology, then make it survivable++The patterns aren't exotic — they're the small vocabulary of multi-agent design, and most systems are one+of them (or a couple composed). The skill is matching the topology to the problem's dependency structure,+resisting the urge to over-architect, and then hardening whatever you chose against its specific failure+mode. A pattern without resilience is the project that fails in month three.++Want to watch the supervisor pattern run locally — a plain-language orchestrator decomposing your intent+across a team you can actually see? You can [download Munder Difflin](/#install) free; it's open source.diff --git a/docs/blog/claude-code-git-worktrees-vs-hive/index.html b/docs/blog/claude-code-git-worktrees-vs-hive/index.htmlindex 0870aa4b..b99c1bd3 100644--- a/docs/blog/claude-code-git-worktrees-vs-hive/index.html+++ b/docs/blog/claude-code-git-worktrees-vs-hive/index.html@@ -291,56 +291,56 @@ team; it’s free and open source.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>- <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta"><time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>6 min</span>+ <span>5 min</span></div>- <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>- <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/human-in-the-loop-ai-agents/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-02">Jun 2, 2026</time>+ <time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>5 min</span>+ <span>6 min</span></div>- <h3><a href="/blog/human-in-the-loop-ai-agents/">Human-in-the-Loop AI Agents: Approvals That Don't Slow You Down</a></h3>- <p class="dek">Design human-in-the-loop AI agents that stay autonomous: an approvals queue that escalates only spend, destructive ops, and scope changes.</p>+ <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>+ <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/human-in-the-loop-ai-agents/" aria-label="Read: Human-in-the-Loop AI Agents: Approvals That Don't Slow You Down">Read →</a>+ <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a></div></div></article>diff --git a/docs/blog/claude-code-orchestration-guide/index.html b/docs/blog/claude-code-orchestration-guide/index.htmlindex 76e578b0..d7214968 100644--- a/docs/blog/claude-code-orchestration-guide/index.html+++ b/docs/blog/claude-code-orchestration-guide/index.html@@ -289,25 +289,25 @@ agents; it’s free and open source.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>- <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body">@@ -316,29 +316,29 @@ agents; it’s free and open source.</p><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>- <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta"><time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>6 min</span>+ <span>5 min</span></div>- <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>- <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a></div></div></article>diff --git a/docs/blog/coordinating-ai-coding-agents/index.html b/docs/blog/coordinating-ai-coding-agents/index.htmlindex 9c4f1463..12903820 100644--- a/docs/blog/coordinating-ai-coding-agents/index.html+++ b/docs/blog/coordinating-ai-coding-agents/index.html@@ -276,25 +276,25 @@ your own team of Claude Code agents.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>- <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body">@@ -303,29 +303,29 @@ your own team of Claude Code agents.</p><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>- <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta"><time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>6 min</span>+ <span>5 min</span></div>- <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>- <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a></div></div></article>diff --git a/docs/blog/feed.xml b/docs/blog/feed.xmlindex 51caaa19..c8edcfbb 100644--- a/docs/blog/feed.xml+++ b/docs/blog/feed.xml@@ -47,6 +47,14 @@<updated>2026-06-05T00:00:00.000Z</updated><category term="Guides" /><summary>How the Model Context Protocol gets attacked — tool poisoning, rug pulls, command injection — and the layered defenses that secure your MCP tools.</summary></entry>+ <entry>+ <title>Multi-Agent Orchestration Patterns: Which Topology, and When</title>+ <link href="https://munderdiffl.in/blog/multi-agent-orchestration-patterns/" />+ <id>https://munderdiffl.in/blog/multi-agent-orchestration-patterns/</id>+ <published>2026-06-05T00:00:00.000Z</published>+ <updated>2026-06-05T00:00:00.000Z</updated><category term="Orchestration" />+ <summary>The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</summary>+ </entry><entry><title>The Multi-Agent Cost Playbook: Five Levers to Cut AI Agent Spend</title><link href="https://munderdiffl.in/blog/the-multi-agent-cost-playbook/" />@@ -479,12 +487,4 @@<updated>2026-06-02T00:00:00.000Z</updated><category term="Concepts" /><summary>Answers to the top Munder Difflin questions — what it is, is it free, does it run locally, which platforms, and how it differs from many terminals.</summary></entry>- <entry>- <title>Run an Office of AI Agents While You Sleep</title>- <link href="https://munderdiffl.in/blog/run-an-office-of-ai-agents/" />- <id>https://munderdiffl.in/blog/run-an-office-of-ai-agents/</id>- <published>2026-06-02T00:00:00.000Z</published>- <updated>2026-06-02T00:00:00.000Z</updated><category term="Use Cases" />- <summary>What it takes to run a self-coordinating office of AI coding agents that keeps shipping after you log off — and the guardrails that keep it sane.</summary>- </entry></feed>diff --git a/docs/blog/how-the-god-orchestrator-works/index.html b/docs/blog/how-the-god-orchestrator-works/index.htmlindex 605c28e4..af99c816 100644--- a/docs/blog/how-the-god-orchestrator-works/index.html+++ b/docs/blog/how-the-god-orchestrator-works/index.html@@ -318,25 +318,25 @@ to put one in charge of your floor; it’s free and open source.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>- <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body">@@ -345,29 +345,29 @@ to put one in charge of your floor; it’s free and open source.</p><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>- <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta"><time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>6 min</span>+ <span>5 min</span></div>- <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>- <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a></div></div></article>diff --git a/docs/blog/human-in-the-loop-ai-agents/index.html b/docs/blog/human-in-the-loop-ai-agents/index.htmlindex 6526a686..8c16a83a 100644--- a/docs/blog/human-in-the-loop-ai-agents/index.html+++ b/docs/blog/human-in-the-loop-ai-agents/index.html@@ -279,25 +279,25 @@ and ask only when they should; it’s free and open source.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>- <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body">@@ -306,29 +306,29 @@ and ask only when they should; it’s free and open source.</p><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>- <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta"><time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>6 min</span>+ <span>5 min</span></div>- <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>- <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a></div></div></article>diff --git a/docs/blog/index.html b/docs/blog/index.htmlindex 39588ca4..fcb06326 100644--- a/docs/blog/index.html+++ b/docs/blog/index.html@@ -85,7 +85,7 @@building a self-coordinating hive of agents with a GOD orchestrator you talk to.</p><div class="filters" role="navigation" aria-label="Topics">- <a class="chip" aria-current="page">All<span class="ct">78</span></a>+ <a class="chip" aria-current="page">All<span class="ct">79</span></a><a class="chip" href="/blog/topics/guides/">guides<span class="ct">21</span></a>@@ -95,7 +95,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>@@ -203,6 +203,24 @@</div></div></article>+<article class="card">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a>+ </div>+ </div>+</article><article class="card"><a class="thumb t-guides" href="/blog/the-multi-agent-cost-playbook/" aria-hidden="true" tabindex="-1"><span>GUIDES</span>diff --git a/docs/blog/local-first-ai-agent-orchestration/index.html b/docs/blog/local-first-ai-agent-orchestration/index.htmlindex adb65007..d658bf97 100644--- a/docs/blog/local-first-ai-agent-orchestration/index.html+++ b/docs/blog/local-first-ai-agent-orchestration/index.html@@ -286,56 +286,56 @@ loop; it’s free and open source.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>- <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta"><time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>6 min</span>+ <span>5 min</span></div>- <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>- <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/human-in-the-loop-ai-agents/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-02">Jun 2, 2026</time>+ <time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span>- <span>5 min</span>+ <span>6 min</span></div>- <h3><a href="/blog/human-in-the-loop-ai-agents/">Human-in-the-Loop AI Agents: Approvals That Don't Slow You Down</a></h3>- <p class="dek">Design human-in-the-loop AI agents that stay autonomous: an approvals queue that escalates only spend, destructive ops, and scope changes.</p>+ <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>+ <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/human-in-the-loop-ai-agents/" aria-label="Read: Human-in-the-Loop AI Agents: Approvals That Don't Slow You Down">Read →</a>+ <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a></div></div></article>diff --git a/docs/blog/mcp-security-tool-poisoning/index.html b/docs/blog/mcp-security-tool-poisoning/index.htmlindex 3daf3865..fa41700a 100644--- a/docs/blog/mcp-security-tool-poisoning/index.html+++ b/docs/blog/mcp-security-tool-poisoning/index.html@@ -266,7 +266,7 @@ execution. Do that and MCP is what it should be: a powerful, <em>bounded</em> wa</div><footer class="post-foot wrap"><div class="prevnext">- <a class="pn prev" href="/blog/the-multi-agent-cost-playbook/"><span class="k">← Older</span><span class="t">The Multi-Agent Cost Playbook: Five Levers to Cut AI Agent Spend</span></a>+ <a class="pn prev" href="/blog/multi-agent-orchestration-patterns/"><span class="k">← Older</span><span class="t">Multi-Agent Orchestration Patterns: Which Topology, and When</span></a><a class="pn next" href="/blog/how-to-install-and-use-munder-difflin/"><span class="k">Newer →</span><span class="t">How to Install and Use Munder Difflin</span></a></div><section class="related">diff --git a/docs/blog/multi-agent-orchestration-patterns/index.html b/docs/blog/multi-agent-orchestration-patterns/index.htmlnew file mode 100644index 00000000..54f7d18b--- /dev/null+++ b/docs/blog/multi-agent-orchestration-patterns/index.html@@ -0,0 +1,383 @@+<!doctype html>+<html lang="en">+<head>+<meta charset="UTF-8" />+<meta name="viewport" content="width=device-width, initial-scale=1.0" />+<title>Multi-Agent Orchestration Patterns: Which Topology, and When — Munder Difflin Blog</title>+<meta name="description" content="The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins." />+<link rel="canonical" href="https://munderdiffl.in/blog/multi-agent-orchestration-patterns/" />++<link rel="icon" type="image/png" href="https://munderdiffl.in/logo.png" />+<link rel="apple-touch-icon" href="https://munderdiffl.in/logo.png" />+<meta name="theme-color" content="#F5F2E8" />++<!-- Open Graph -->+<meta property="og:site_name" content="Munder Difflin Blog" />+<meta property="og:type" content="article" />+<meta property="og:title" content="Multi-Agent Orchestration Patterns: Which Topology, and When — Munder Difflin Blog" />+<meta property="og:description" content="The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins." />+<meta property="og:url" content="https://munderdiffl.in/blog/multi-agent-orchestration-patterns/" />+<meta property="og:image" content="https://munderdiffl.in/media/og.png" />+<meta property="og:locale" content="en_US" />+<meta property="article:published_time" content="2026-06-05T00:00:00.000Z" />+<meta property="article:author" content="Chaitanya Giri" />+<meta property="article:tag" content="Orchestration" />+<meta property="article:tag" content="Multi-Agent" />+<meta property="article:tag" content="Concepts" />+<meta property="article:tag" content="Architecture" />+++<!-- Twitter -->+<meta name="twitter:card" content="summary_large_image" />+<meta name="twitter:title" content="Multi-Agent Orchestration Patterns: Which Topology, and When — Munder Difflin Blog" />+<meta name="twitter:description" content="The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins." />+<meta name="twitter:image" content="https://munderdiffl.in/media/og.png" />++<link rel="alternate" type="application/atom+xml" title="Munder Difflin Blog" href="https://munderdiffl.in/blog/feed.xml" />+<link rel="preconnect" href="https://fonts.googleapis.com" />+<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />+<link href="https://fonts.googleapis.com/css2?family=Geist:wght@400;500;600&family=JetBrains+Mono:wght@400;500;600;700&display=swap" rel="stylesheet" />+<link rel="stylesheet" href="/blog/assets/blog.css" />++<script type="application/ld+json">+{+ "@context": "https://schema.org",+ "@type": "BlogPosting",+ "headline": "Multi-Agent Orchestration Patterns: Which Topology, and When",+ "description": "The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.",+ "image": ["https://munderdiffl.in/media/og.png"],+ "datePublished": "2026-06-05T00:00:00.000Z",+ "dateModified": "2026-06-05T00:00:00.000Z",+ "author": { "@type": "Person", "name": "Chaitanya Giri" },+ "publisher": {+ "@type": "Organization",+ "name": "Munder Difflin",+ "logo": { "@type": "ImageObject", "url": "https://munderdiffl.in/logo.png" }+ },+ "mainEntityOfPage": { "@type": "WebPage", "@id": "https://munderdiffl.in/blog/multi-agent-orchestration-patterns/" },+ "keywords": "multi-agent orchestration patterns, agent orchestration topologies, orchestrator worker pattern, supervisor agent pattern, when to use multi-agent",+ "articleSection": "Orchestration",+ "inLanguage": "en-US",+ "url": "https://munderdiffl.in/blog/multi-agent-orchestration-patterns/"+}+</script>+<script type="application/ld+json">+{+ "@context": "https://schema.org",+ "@type": "BreadcrumbList",+ "itemListElement": [+ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://munderdiffl.in/" },+ { "@type": "ListItem", "position": 2, "name": "Blog", "item": "https://munderdiffl.in/blog/" },+ { "@type": "ListItem", "position": 3, "name": "Multi-Agent Orchestration Patterns: Which Topology, and When", "item": "https://munderdiffl.in/blog/multi-agent-orchestration-patterns/" }+ ]+}+</script>++<script type="application/ld+json">+{+ "@context": "https://schema.org",+ "@type": "FAQPage",+ "mainEntity": [+ { "@type": "Question", "name": "What are the main multi-agent orchestration patterns?", "acceptedAnswer": { "@type": "Answer", "text": "Six recur: orchestrator-worker (a.k.a. supervisor — a central agent decomposes and delegates), hierarchical (supervisors of supervisors), sequential pipeline (fixed linear stages), parallel fan-out (scatter-gather over independent tasks), debate/maker-checker (one agent produces, another verifies), and blackboard (agents coordinate through shared state). Most production systems lean on orchestrator-worker." } },+ { "@type": "Question", "name": "Which orchestration pattern should I use?", "acceptedAnswer": { "@type": "Answer", "text": "Match it to your dependency structure: linear dependencies → pipeline; independent tasks → fan-out; accuracy-critical work → maker-checker; known sub-tasks with one accountable owner → orchestrator-worker (the 2026 default). Start with the simplest pattern that fits — most teams over-architect and add agents they don't need." } },+ { "@type": "Question", "name": "Is a multi-agent system always better than a single agent?", "acceptedAnswer": { "@type": "Answer", "text": "No. In benchmarks, a single agent matched or beat multi-agent systems on a majority of tasks when given the same tools and context, and a large share of multi-agent pilots fail in production — usually from the wrong pattern. More agents add coordination cost; they pay off only when the task genuinely decomposes." } }++ ]+}+</script>+++</head>+<body>+<a class="skip-link" href="#main">Skip to content</a>+<nav id="nav">+ <a class="brand" href="/blog/" aria-label="Munder Difflin Blog home">+ <span class="brand-mark">MD</span>+ <span class="brand-word">MUNDER DIFFLIN<span class="sep">/</span><span class="blog">BLOG</span></span>+ </a>+ <span class="spacer"></span>+ <div class="links">+ <a href="/blog/">Latest</a>+ <a href="/blog/topics/">Topics</a>+ <a href="/blog/about/">About</a>+ <a href="https://munderdiffl.in">↗ Product</a>+ </div>+ <div class="nav-actions">+ <a class="btn sm" href="https://munderdiffl.in/blog/feed.xml" aria-label="RSS feed">⤵ RSS</a>+ <a class="btn sm primary" href="https://munderdiffl.in/#install">⤓ Download</a>+ </div>+</nav>++<main id="main">++<article>+ <header class="post-head wrap">+ <nav class="crumbs" aria-label="Breadcrumb">+ <a href="/blog/">Blog</a>+ <span>/</span>+ <a href="/blog/topics/orchestration/">Orchestration</a>+ </nav>+ <h1>Multi-Agent Orchestration Patterns: Which Topology, and When</h1>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="byline">+ <span class="av" aria-hidden="true">CG</span>+ <strong>Chaitanya Giri</strong>+ <span class="dot" aria-hidden="true">·</span>+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true">·</span>+ <span>5 min read</span>+ <span class="tag kind">Orchestration</span>+ </div>+ </header>++ <div class="post-wrap wrap">+ <div class="prose">+ <div class="callout tldr"><span class="ic">TL;DR</span><p>Multi-agent systems come in a few named+<strong>topologies</strong> — orchestrator-worker (supervisor), pipeline, fan-out, debate/maker-checker,+swarm, and blackboard. <strong>Orchestrator-worker is the 2026 default.</strong> But the most important+fact is the anti-hype one: a <strong>single agent often matches or beats</strong> a multi-agent system on+the same task, and many multi-agent pilots fail — usually from picking the wrong pattern. Rule: start with+the simplest topology that fits, add agents only when the problem genuinely decomposes, and make whatever+you pick <strong>survivable</strong>.</p></div>+<p>Once you’ve decided you need more than one agent, the next question is <em>how they’re wired together</em> —+the topology. There’s a small, well-worn set of patterns, each suited to a different shape of problem.+This is the catalog, when to reach for each, the honest caveats, and how the patterns show up in a real+<a href="/#how">multi-agent hive</a>.</p>+<h2 id="the-pattern-catalog" tabindex="-1">The pattern catalog <a class="anchor" href="#the-pattern-catalog" aria-hidden="true">#</a></h2>+<ul>+<li><strong>Orchestrator-worker (supervisor).</strong> A central orchestrator classifies the task, decomposes it,+dispatches sub-tasks to specialized workers, and merges the results. One clear accountability point, and+workers can run cheaper task-specific models — say, a research orchestrator that dispatches lookups to+small, fast workers and reserves a strong model for the final synthesis. <strong>Use when</strong> you know the+sub-tasks at design time and want one place that owns the outcome. This is the 2026 production workhorse.</li>+<li><strong>Hierarchical (supervisors of supervisors).</strong> Tiered orchestrator-worker: higher tiers plan and+coordinate, lower tiers execute. <strong>Use when</strong> the work is too big for one orchestrator’s context — teams+of teams.</li>+<li><strong>Sequential pipeline.</strong> Fixed linear stages, each consuming the previous output. <strong>Use when</strong> the order+never changes and steps strictly depend on each other (extract → transform → validate).</li>+<li><strong>Parallel fan-out (scatter-gather).</strong> Dispatch N independent sub-tasks at once, then gather. <strong>Use+when</strong> you have several tasks with no dependencies between them — the classic latency win.</li>+<li><strong>Debate / maker-checker.</strong> One agent produces, another critiques or verifies; iterate or vote. <strong>Use+when</strong> accuracy matters more than speed — review loops, adversarial checks.</li>+<li><strong>Swarm / network (peer).</strong> Dynamic peer agents hand off to each other with no central conductor. <strong>Use+when</strong> the path is emergent and data-dependent — and budget for it being the hardest to debug and bound.</li>+<li><strong>Blackboard.</strong> Agents read and write a shared workspace rather than messaging directly; the shared+state coordinates them. <strong>Use when</strong> many agents contribute to one evolving artifact.</li>+</ul>+<h2 id="a-decision-framework" tabindex="-1">A decision framework <a class="anchor" href="#a-decision-framework" aria-hidden="true">#</a></h2>+<p>You can get most of the way with three rules:</p>+<ol>+<li><strong>Default to supervisor / orchestrator-worker.</strong> It’s legible, accountable, and debuggable — which is+why it dominates production.</li>+<li><strong>Match the topology to the dependency structure.</strong> Linear deps → pipeline. No deps → fan-out.+Accuracy-critical → maker-checker. Emergent path → swarm (with guardrails).</li>+<li><strong>Start simple.</strong> Most teams over-architect. Reach for more agents only when a single agent genuinely+can’t hold the task.</li>+</ol>+<h2 id="patterns-compose" tabindex="-1">Patterns compose <a class="anchor" href="#patterns-compose" aria-hidden="true">#</a></h2>+<p>Real systems rarely use one pattern in isolation — they nest. A supervisor’s workers might each run a+maker-checker loop internally; a sequential pipeline might fan out within a single stage; a hierarchical+team might keep a blackboard at each tier to pool findings. The patterns are a <em>vocabulary</em>, not a menu you+pick one item from. The discipline is the same at every level of nesting: at each composition point, ask+what the dependency structure of <em>that</em> sub-problem is, and choose the topology that fits it.</p>+<p>A common, effective shape is a <strong>supervisor on the outside</strong> — one accountable owner with clear+decomposition — wrapping <strong>fan-out and maker-checker on the inside</strong>: run the independent sub-tasks in+parallel, and route the risky or accuracy-critical ones through a verifier. Compose deliberately, though:+every boundary you add is another place coordination can fail, and another handoff to debug.</p>+<h2 id="the-honest-part-multi-agent-isnt-free" tabindex="-1">The honest part: multi-agent isn’t free <a class="anchor" href="#the-honest-part-multi-agent-isnt-free" aria-hidden="true">#</a></h2>+<p>Two findings are worth leading with, so this stays grounded and not hype:</p>+<ul>+<li><strong>A single agent often wins.</strong> In benchmarking, a single agent matched or outperformed multi-agent+systems on a majority of tasks when given the same tools and context. More agents add coordination+overhead, latency, and cost — the payoff only appears when the task truly decomposes.</li>+<li><strong>Many pilots fail.</strong> A large share of multi-agent projects don’t survive their first months in+production, commonly because the team picked the wrong pattern, or the right pattern without+understanding its failure mode — a swarm with no hop cap loops; a fan-out with hidden cross-task+dependencies corrupts its merge.</li>+</ul>+<p>The corollary: the value isn’t “more agents,” it’s <strong>the right topology plus the resilience to survive+its failure mode</strong> — hop caps, idempotent handling, isolation. (We wrote up those mechanics in+<a href="/blog/recovering-from-agent-failures/">recovering from agent failures</a>.)</p>+<h2 id="the-2026-framework-landscape" tabindex="-1">The 2026 framework landscape <a class="anchor" href="#the-2026-framework-landscape" aria-hidden="true">#</a></h2>+<p>If you’re building this on a framework, the control-vs-simplicity tension hasn’t converged — they differ+mostly in orchestration model and ecosystem:</p>+<ul>+<li><strong>LangGraph</strong> — a directed graph with conditional edges; the most control, the most boilerplate. The+default for stateful, auditable, regulated workflows.</li>+<li><strong>CrewAI</strong> — role-based “crews” with process types; the fastest path to a working multi-agent team.</li>+<li><strong>AutoGen / AG2</strong> — conversational group chat between agents.</li>+<li><strong>OpenAI Agents SDK</strong> — explicit handoffs between agents, with a built-in execution sandbox.</li>+<li><strong>Claude Agent SDK / Strands</strong> — simplicity over fine-grained orchestration.</li>+</ul>+<p>The 2026 trend is convergence on common abstractions (graphs, roles, handoffs) with differentiation in+ecosystem depth and ops tooling. Pick for the orchestration model you need, not the logo.</p>+<h2 id="how-the-patterns-map-onto-a-hive" tabindex="-1">How the patterns map onto a hive <a class="anchor" href="#how-the-patterns-map-onto-a-hive" aria-hidden="true">#</a></h2>+<p>A concrete system makes the patterns less abstract. Munder Difflin runs several at once:</p>+<ul>+<li><strong>Orchestrator-worker / supervisor</strong> — the <a href="/blog/how-the-god-orchestrator-works/">god orchestrator</a>+classifies intent, decomposes it, dispatches to role-specialized agents, and owns integration. That’s+the 2026 default, made local and watchable.</li>+<li><strong>Blackboard</strong> — the shared board and the append-only event log are a blackboard the team reads and+writes.</li>+<li><strong>Fan-out</strong> — a broadcast message is scatter-gather over the active roster.</li>+<li><strong>Maker-checker</strong> — human-in-the-loop approval and adversarial review act as a consensus gate on risky+output.</li>+</ul>+<p>It’s the same lesson throughout: orchestration is how agents <a href="/blog/coordinating-ai-coding-agents/">coordinate without+colliding</a>, and the pattern you choose is a big part of whether they+do.</p>+<h2 id="pick-the-topology-then-make-it-survivable" tabindex="-1">Pick the topology, then make it survivable <a class="anchor" href="#pick-the-topology-then-make-it-survivable" aria-hidden="true">#</a></h2>+<p>The patterns aren’t exotic — they’re the small vocabulary of multi-agent design, and most systems are one+of them (or a couple composed). The skill is matching the topology to the problem’s dependency structure,+resisting the urge to over-architect, and then hardening whatever you chose against its specific failure+mode. A pattern without resilience is the project that fails in month three.</p>+<p>Want to watch the supervisor pattern run locally — a plain-language orchestrator decomposing your intent+across a team you can actually see? You can <a href="/#install">download Munder Difflin</a> free; it’s open source.</p>++++ <section class="faq" aria-label="Frequently asked questions">+ <h2 id="faq">FAQ</h2>++ <div class="qa">+ <p class="q">What are the main multi-agent orchestration patterns?</p>+ <p class="a">Six recur: orchestrator-worker (a.k.a. supervisor — a central agent decomposes and delegates), hierarchical (supervisors of supervisors), sequential pipeline (fixed linear stages), parallel fan-out (scatter-gather over independent tasks), debate/maker-checker (one agent produces, another verifies), and blackboard (agents coordinate through shared state). Most production systems lean on orchestrator-worker.</p>+ </div>++ <div class="qa">+ <p class="q">Which orchestration pattern should I use?</p>+ <p class="a">Match it to your dependency structure: linear dependencies → pipeline; independent tasks → fan-out; accuracy-critical work → maker-checker; known sub-tasks with one accountable owner → orchestrator-worker (the 2026 default). Start with the simplest pattern that fits — most teams over-architect and add agents they don't need.</p>+ </div>++ <div class="qa">+ <p class="q">Is a multi-agent system always better than a single agent?</p>+ <p class="a">No. In benchmarks, a single agent matched or beat multi-agent systems on a majority of tasks when given the same tools and context, and a large share of multi-agent pilots fail in production — usually from the wrong pattern. More agents add coordination cost; they pay off only when the task genuinely decomposes.</p>+ </div>++ </section>++++ <hr />+ <div class="share">+ <span class="lbl">Tags</span>+ <a class="tag" href="/blog/tags/orchestration/">Orchestration</a><a class="tag" href="/blog/tags/multi-agent/">Multi-Agent</a><a class="tag" href="/blog/tags/concepts/">Concepts</a><a class="tag" href="/blog/tags/architecture/">Architecture</a>+ </div>++ </div>+++ <aside class="toc-rail" aria-label="Table of contents">+ <div class="tt">On this page</div>+ <ol>+ <li class="lvl-2"><a href="#the-pattern-catalog">The pattern catalog</a></li><li class="lvl-2"><a href="#a-decision-framework">A decision framework</a></li><li class="lvl-2"><a href="#patterns-compose">Patterns compose</a></li><li class="lvl-2"><a href="#the-honest-part-multi-agent-isnt-free">The honest part: multi-agent isn’t free</a></li><li class="lvl-2"><a href="#the-2026-framework-landscape">The 2026 framework landscape</a></li><li class="lvl-2"><a href="#how-the-patterns-map-onto-a-hive">How the patterns map onto a hive</a></li><li class="lvl-2"><a href="#pick-the-topology-then-make-it-survivable">Pick the topology, then make it survivable</a></li>+ </ol>+ </aside>++ </div>++ <footer class="post-foot wrap"><div class="prevnext">+ <a class="pn prev" href="/blog/the-multi-agent-cost-playbook/"><span class="k">← Older</span><span class="t">The Multi-Agent Cost Playbook: Five Levers to Cut AI Agent Spend</span></a>+ <a class="pn next" href="/blog/mcp-security-tool-poisoning/"><span class="k">Newer →</span><span class="t">MCP Tool Poisoning: How to Secure the Model Context Protocol</span></a>+ </div>+ <section class="related">+ <h2>Related in Orchestration</h2>+ <div class="post-grid">+ <article class="card">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-04">Jun 4, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ </div>+ </div>+</article>+<article class="card">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-04">Jun 4, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ </div>+ </div>+</article>+<article class="card">+ <a class="thumb t-orchestration" href="/blog/scheduling-autonomous-agent-missions/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-04">Jun 4, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>6 min</span>+ </div>+ <h3><a href="/blog/scheduling-autonomous-agent-missions/">Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions</a></h3>+ <p class="dek">How scheduled missions make a Claude Code hive work on a cadence — recurring requests that fire on an interval and land in an agent's queue, hands-free.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/scheduling-autonomous-agent-missions/" aria-label="Read: Make a Hive Work While You Sleep: Scheduling Autonomous Agent Missions">Read →</a>+ </div>+ </div>+</article>++ </div>+ </section>++ </footer>+</article>++</main>+<footer>+ <div class="wrap">+ <div class="foot-top">+ <a class="foot-brand brand" href="/blog/">+ <span class="brand-mark">MD</span>+ <span class="brand-word">MUNDER DIFFLIN<span class="sep">/</span><span class="blog">BLOG</span></span>+ </a>+ <div class="foot-links">+ <a href="/blog/">Latest</a>+ <a href="/blog/topics/">Topics</a>+ <a href="/blog/about/">About</a>+ <a href="https://munderdiffl.in/blog/feed.xml">RSS</a>+ <a href="https://github.com/chaitanyagiri/munder-difflin">GitHub</a>+ <a href="https://munderdiffl.in">munderdiffl.in ↗</a>+ </div>+ </div>+ <div class="foot-meta">+ © 2026 Munder Difflin · Local-first · MIT-licensed ·+ Built on <a href="https://munderdiffl.in">Electron, React, Pixi.js, xterm.js & node-pty</a>.+ Run an office of agents while you sleep.+ </div>+ </div>+</footer>++<script>+ // sticky-nav border on scroll+ (function () {+ var nav = document.getElementById('nav');+ if (!nav) return;+ var onScroll = function () { nav.classList.toggle('scrolled', window.scrollY > 8); };+ onScroll(); window.addEventListener('scroll', onScroll, { passive: true });+ })();+</script>++</body>+</html>diff --git a/docs/blog/scheduling-autonomous-agent-missions/index.html b/docs/blog/scheduling-autonomous-agent-missions/index.htmlindex cfa89004..de625130 100644--- a/docs/blog/scheduling-autonomous-agent-missions/index.html+++ b/docs/blog/scheduling-autonomous-agent-missions/index.html@@ -297,25 +297,25 @@ mission and watch the floor wake itself up; it’s free and open source.</p><h2>Related in Orchestration</h2><div class="post-grid"><article class="card">- <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-04">Jun 4, 2026</time>+ <time datetime="2026-06-05">Jun 5, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>- <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body">@@ -324,29 +324,29 @@ mission and watch the floor wake itself up; it’s free and open source.</p><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>- <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p>+ <h3><a href="/blog/claude-code-git-worktrees-vs-hive/">Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code</a></h3>+ <p class="dek">Claude Code git worktrees vs a hive: when isolated worktrees suffice for parallel agents, and when you need memory, messaging, and an orchestrator on top.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a>+ <a class="more" href="/blog/claude-code-git-worktrees-vs-hive/" aria-label="Read: Git Worktrees vs a Hive: Two Ways to Parallelize Claude Code">Read →</a></div></div></article><article class="card">- <a class="thumb t-orchestration" href="/blog/human-in-the-loop-ai-agents/" aria-hidden="true" tabindex="-1">+ <a class="thumb t-orchestration" href="/blog/local-first-ai-agent-orchestration/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a><div class="body"><div class="meta">- <time datetime="2026-06-02">Jun 2, 2026</time>+ <time datetime="2026-06-04">Jun 4, 2026</time><span class="dot" aria-hidden="true"></span><span>5 min</span></div>- <h3><a href="/blog/human-in-the-loop-ai-agents/">Human-in-the-Loop AI Agents: Approvals That Don't Slow You Down</a></h3>- <p class="dek">Design human-in-the-loop AI agents that stay autonomous: an approvals queue that escalates only spend, destructive ops, and scope changes.</p>+ <h3><a href="/blog/local-first-ai-agent-orchestration/">Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud</a></h3>+ <p class="dek">How local-first AI agent orchestration works under the hood — the loop, mailboxes, scheduler, and audit log that coordinate a hive on one machine.</p><div class="foot"><span class="tag kind">Orchestration</span>- <a class="more" href="/blog/human-in-the-loop-ai-agents/" aria-label="Read: Human-in-the-Loop AI Agents: Approvals That Don't Slow You Down">Read →</a>+ <a class="more" href="/blog/local-first-ai-agent-orchestration/" aria-label="Read: Local-First AI Agent Orchestration: How a Hive Runs Without a Cloud">Read →</a></div></div></article>diff --git a/docs/blog/tags/architecture/index.html b/docs/blog/tags/architecture/index.htmlindex d44e949f..4599cd6a 100644--- a/docs/blog/tags/architecture/index.html+++ b/docs/blog/tags/architecture/index.html@@ -58,11 +58,29 @@<header class="page-head wrap"><p class="eyebrow">Tag</p><h1>#Architecture</h1>- <p class="lead">2 posts tagged <strong>Architecture</strong>.</p>+ <p class="lead">3 posts tagged <strong>Architecture</strong>.</p></header><section class="wrap"><div class="post-grid"><article class="card">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a>+ </div>+ </div>+</article>+<article class="card"><a class="thumb t-internals" href="/blog/architecture-two-planes-one-renderer/" aria-hidden="true" tabindex="-1"><span>INTERNALS</span></a>diff --git a/docs/blog/tags/concepts/index.html b/docs/blog/tags/concepts/index.htmlindex b3d58c8e..619d8ff2 100644--- a/docs/blog/tags/concepts/index.html+++ b/docs/blog/tags/concepts/index.html@@ -58,7 +58,7 @@<header class="page-head wrap"><p class="eyebrow">Tag</p><h1>#Concepts</h1>- <p class="lead">10 posts tagged <strong>Concepts</strong>.</p>+ <p class="lead">11 posts tagged <strong>Concepts</strong>.</p></header><section class="wrap"><div class="post-grid">@@ -80,6 +80,24 @@</div></div></article>+<article class="card">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a>+ </div>+ </div>+</article><article class="card"><a class="thumb t-concepts" href="/blog/agents-rule-of-two/" aria-hidden="true" tabindex="-1"><span>CONCEPTS</span>diff --git a/docs/blog/tags/multi-agent/index.html b/docs/blog/tags/multi-agent/index.htmlindex 130a3d45..7017b2b0 100644--- a/docs/blog/tags/multi-agent/index.html+++ b/docs/blog/tags/multi-agent/index.html@@ -58,7 +58,7 @@<header class="page-head wrap"><p class="eyebrow">Tag</p><h1>#Multi-Agent</h1>- <p class="lead">53 posts tagged <strong>Multi-Agent</strong>.</p>+ <p class="lead">54 posts tagged <strong>Multi-Agent</strong>.</p></header><section class="wrap"><div class="post-grid">@@ -98,6 +98,24 @@</div></div></article>+<article class="card">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a>+ </div>+ </div>+</article><article class="card"><a class="thumb t-guides" href="/blog/the-multi-agent-cost-playbook/" aria-hidden="true" tabindex="-1"><span>GUIDES</span>diff --git a/docs/blog/tags/orchestration/index.html b/docs/blog/tags/orchestration/index.htmlindex 72149572..7ff21b1c 100644--- a/docs/blog/tags/orchestration/index.html+++ b/docs/blog/tags/orchestration/index.html@@ -58,11 +58,29 @@<header class="page-head wrap"><p class="eyebrow">Tag</p><h1>#Orchestration</h1>- <p class="lead">12 posts tagged <strong>Orchestration</strong>.</p>+ <p class="lead">13 posts tagged <strong>Orchestration</strong>.</p></header><section class="wrap"><div class="post-grid"><article class="card">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a>+ </div>+ </div>+</article>+<article class="card"><a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a>diff --git a/docs/blog/the-multi-agent-cost-playbook/index.html b/docs/blog/the-multi-agent-cost-playbook/index.htmlindex 26013f6a..fc4165aa 100644--- a/docs/blog/the-multi-agent-cost-playbook/index.html+++ b/docs/blog/the-multi-agent-cost-playbook/index.html@@ -266,7 +266,7 @@ to run a fleet that doesn’t bankrupt you — it’s free and open source.</p><footer class="post-foot wrap"><div class="prevnext"><a class="pn prev" href="/blog/agent-security-and-sandboxing/"><span class="k">← Older</span><span class="t">Running AI Agents Safely: Permission Modes and Sandboxing</span></a>- <a class="pn next" href="/blog/mcp-security-tool-poisoning/"><span class="k">Newer →</span><span class="t">MCP Tool Poisoning: How to Secure the Model Context Protocol</span></a>+ <a class="pn next" href="/blog/multi-agent-orchestration-patterns/"><span class="k">Newer →</span><span class="t">Multi-Agent Orchestration Patterns: Which Topology, and When</span></a></div><section class="related"><h2>Related in Guides</h2>diff --git a/docs/blog/topics/comparisons/index.html b/docs/blog/topics/comparisons/index.htmlindex efb425e5..ab391a01 100644--- a/docs/blog/topics/comparisons/index.html+++ b/docs/blog/topics/comparisons/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/" aria-current="page">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>diff --git a/docs/blog/topics/concepts/index.html b/docs/blog/topics/concepts/index.htmlindex 3c0dc95b..7ac91822 100644--- a/docs/blog/topics/concepts/index.html+++ b/docs/blog/topics/concepts/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>diff --git a/docs/blog/topics/guides/index.html b/docs/blog/topics/guides/index.htmlindex afbdf480..da29aabc 100644--- a/docs/blog/topics/guides/index.html+++ b/docs/blog/topics/guides/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>diff --git a/docs/blog/topics/index.html b/docs/blog/topics/index.htmlindex 95b70326..da5b473a 100644--- a/docs/blog/topics/index.html+++ b/docs/blog/topics/index.html@@ -106,7 +106,7 @@<span class="thumb t-orchestration" aria-hidden="true">ORCHESTRATION</span><div class="body"><h3 style="color:var(--ink)">orchestration</h3>- <p class="dek">7 posts.</p>+ <p class="dek">8 posts.</p><div class="foot"><span class="more">View topic →</span></div></div></a>@@ -144,20 +144,20 @@<div class="filters" style="padding-top:36px"><span class="eyebrow" style="align-self:center">Tags</span>- <a class="chip" href="/blog/tags/multi-agent/">Multi-Agent<span class="ct">53</span></a>+ <a class="chip" href="/blog/tags/multi-agent/">Multi-Agent<span class="ct">54</span></a><a class="chip" href="/blog/tags/claude-code/">Claude Code<span class="ct">36</span></a><a class="chip" href="/blog/tags/internals/">Internals<span class="ct">21</span></a>- <a class="chip" href="/blog/tags/orchestration/">Orchestration<span class="ct">12</span></a>+ <a class="chip" href="/blog/tags/orchestration/">Orchestration<span class="ct">13</span></a>++ <a class="chip" href="/blog/tags/concepts/">Concepts<span class="ct">11</span></a><a class="chip" href="/blog/tags/guides/">Guides<span class="ct">11</span></a><a class="chip" href="/blog/tags/comparisons/">Comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/tags/concepts/">Concepts<span class="ct">10</span></a>-<a class="chip" href="/blog/tags/memory/">Memory<span class="ct">8</span></a><a class="chip" href="/blog/tags/tools/">Tools<span class="ct">8</span></a>@@ -192,6 +192,8 @@<a class="chip" href="/blog/tags/mcp/">MCP<span class="ct">3</span></a>+ <a class="chip" href="/blog/tags/architecture/">Architecture<span class="ct">3</span></a>+<a class="chip" href="/blog/tags/cost/">Cost<span class="ct">3</span></a><a class="chip" href="/blog/tags/agentic-ai/">Agentic AI<span class="ct">3</span></a>@@ -222,8 +224,6 @@<a class="chip" href="/blog/tags/performance/">Performance<span class="ct">2</span></a>- <a class="chip" href="/blog/tags/architecture/">Architecture<span class="ct">2</span></a>-<a class="chip" href="/blog/tags/reliability/">Reliability<span class="ct">2</span></a><a class="chip" href="/blog/tags/model-routing/">Model Routing<span class="ct">2</span></a>diff --git a/docs/blog/topics/internals/index.html b/docs/blog/topics/internals/index.htmlindex 2b8cb4b9..10b307d2 100644--- a/docs/blog/topics/internals/index.html+++ b/docs/blog/topics/internals/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>diff --git a/docs/blog/topics/memory/index.html b/docs/blog/topics/memory/index.htmlindex dc257de7..097ce649 100644--- a/docs/blog/topics/memory/index.html+++ b/docs/blog/topics/memory/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/" aria-current="page">memory<span class="ct">5</span></a>diff --git a/docs/blog/topics/orchestration/index.html b/docs/blog/topics/orchestration/index.htmlindex cbefb7b2..4dc037e5 100644--- a/docs/blog/topics/orchestration/index.html+++ b/docs/blog/topics/orchestration/index.html@@ -63,7 +63,7 @@</nav><p class="eyebrow">Topic</p><h1>orchestration</h1>- <p class="lead">7 posts in this cluster.</p>+ <p class="lead">8 posts in this cluster.</p><div class="filters" role="navigation" aria-label="Topics"><a class="chip" href="/blog/">All</a>@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/" aria-current="page">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/" aria-current="page">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>@@ -90,6 +90,24 @@<section class="wrap"><div class="post-grid"><article class="card">+ <a class="thumb t-orchestration" href="/blog/multi-agent-orchestration-patterns/" aria-hidden="true" tabindex="-1">+ <span>ORCHESTRATION</span>+ </a>+ <div class="body">+ <div class="meta">+ <time datetime="2026-06-05">Jun 5, 2026</time>+ <span class="dot" aria-hidden="true"></span>+ <span>5 min</span>+ </div>+ <h3><a href="/blog/multi-agent-orchestration-patterns/">Multi-Agent Orchestration Patterns: Which Topology, and When</a></h3>+ <p class="dek">The multi-agent orchestration patterns — orchestrator-worker, pipeline, fan-out, debate, swarm, blackboard — when to use each, and when one agent wins.</p>+ <div class="foot">+ <span class="tag kind">Orchestration</span>+ <a class="more" href="/blog/multi-agent-orchestration-patterns/" aria-label="Read: Multi-Agent Orchestration Patterns: Which Topology, and When">Read →</a>+ </div>+ </div>+</article>+<article class="card"><a class="thumb t-orchestration" href="/blog/claude-code-git-worktrees-vs-hive/" aria-hidden="true" tabindex="-1"><span>ORCHESTRATION</span></a>diff --git a/docs/blog/topics/story/index.html b/docs/blog/topics/story/index.htmlindex 9084bd13..2b30c69d 100644--- a/docs/blog/topics/story/index.html+++ b/docs/blog/topics/story/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>diff --git a/docs/blog/topics/use-cases/index.html b/docs/blog/topics/use-cases/index.htmlindex f88f5068..e0fff8de 100644--- a/docs/blog/topics/use-cases/index.html+++ b/docs/blog/topics/use-cases/index.html@@ -76,7 +76,7 @@<a class="chip" href="/blog/topics/comparisons/">comparisons<span class="ct">11</span></a>- <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">7</span></a>+ <a class="chip" href="/blog/topics/orchestration/">orchestration<span class="ct">8</span></a><a class="chip" href="/blog/topics/memory/">memory<span class="ct">5</span></a>diff --git a/docs/sitemap.xml b/docs/sitemap.xmlindex 2a84991a..697f1bbc 100644--- a/docs/sitemap.xml+++ b/docs/sitemap.xml@@ -47,6 +47,12 @@<changefreq>monthly</changefreq><priority>0.8</priority></url>+ <url>+ <loc>https://munderdiffl.in/blog/multi-agent-orchestration-patterns/</loc>+ <lastmod>2026-06-05</lastmod>+ <changefreq>monthly</changefreq>+ <priority>0.8</priority>+ </url><url><loc>https://munderdiffl.in/blog/the-multi-agent-cost-playbook/</loc><lastmod>2026-06-05</lastmod>@@ -497,9 +503,9 @@<url><loc>https://munderdiffl.in/blog/tags/claude-code/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/internals/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/orchestration/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>+ <url><loc>https://munderdiffl.in/blog/tags/concepts/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/guides/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/comparisons/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>- <url><loc>https://munderdiffl.in/blog/tags/concepts/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/memory/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/tools/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/getting-started/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>@@ -517,6 +523,7 @@<url><loc>https://munderdiffl.in/blog/tags/terminals/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/electron/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/mcp/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>+ <url><loc>https://munderdiffl.in/blog/tags/architecture/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/cost/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/agentic-ai/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/aeo/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>@@ -532,7 +539,6 @@<url><loc>https://munderdiffl.in/blog/tags/xtermjs/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/hooks/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/performance/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>- <url><loc>https://munderdiffl.in/blog/tags/architecture/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/reliability/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/model-routing/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url><url><loc>https://munderdiffl.in/blog/tags/prompt-caching/</loc><changefreq>weekly</changefreq><priority>0.3</priority></url>