Can I customize the agents?

Completely. Each agent is a plain object with a name, role and system instructions. Change the roster, run more in parallel, swap models per agent. It's your code, no config wall.

Runflock — run your own team of AI agents

What you get

A team, not a prompt.

The hard part — running agents in parallel and merging their answers — already done, in code you own.

⚡

Parallel orchestration, not a prompt chain

Your agents run at the same time, not one after another. A lead reads every specialist's output and synthesizes a single, decision-ready answer — the depth of a team with the latency of one call.

🧩

A roster you control

Every agent is a plain object. Rename it, rewrite it, clone it. Want two Researchers and a Security Critic? Change one array.

💬

Terminal or Discord

Run node run.js heads-down, or drop !team in Discord from your phone. Same brain. Bot's in the box.

🛠️

The modern SDK, done right

Streaming and adaptive thinking wired up correctly on the official Anthropic SDK — the parts everyone gets subtly wrong, already handled. Watch the agents think in real time.

📦

~3 files. Yours forever.

No framework to learn, nothing phoning home. Open the orchestrator, read the whole thing in one sitting, change anything. It's your codebase the moment you download it.

Three steps

Running in under five minutes.

Set your key

Drop your Anthropic API key in .env. That's the only account you need.

ANTHROPIC_API_KEY=…

Define your team

Keep the default roster or edit the agents array — names, roles, instructions, however many.

ROLES = [ … ]

Fire a goal

Run it from the terminal or Discord. The team works in parallel; you get one deliverable back.

node run.js "…"

See it run

One goal in. One answer out.

Three specialists, one synthesized deliverable, ~5 seconds of wall-clock time. This is raw output — no editing.

🔊 30-second demo — turn on sound

▶ Pick a goal — watch the team run it

~/runflock — node run.js

Researcheridle

Planneridle

Criticidle

◆ Lead agent — waiting for the specialists…

Curated example runs · the real kit runs your goals from your terminal

Inspect it

The code you own. Read it before you buy.

No black box. This is the actual orchestrator — the ~30 lines that fire every specialist in parallel and merge them into one answer. Copy it, read it: there's no magic, just the fiddly week we already got right.

src/orchestrator.js

// orchestrator.js — every specialist fires at the goal IN PARALLEL,
// then a lead synthesizes their answers into ONE deliverable.
import { runAgent } from "./agent.js";
import { ROLES, SYNTHESIZER } from "./config.js";

export async function orchestrate(goal, { roles = ROLES, onProgress = () => {} } = {}) {
  // 1) every specialist works at the same time
  const specialists = await Promise.all(
    roles.map(async (r) => {
      const text = await runAgent({ system: r.system, goal });
      return { role: r.name, text };
    })
  );

  // 2) the lead merges the specialists into ONE deliverable
  const combined = specialists.map((s) => `## ${s.role}\n${s.text}`).join("\n\n");
  const final = await runAgent({
    system: SYNTHESIZER.system,
    goal: `GOAL: ${goal}\n\nSPECIALIST OUTPUTS:\n${combined}`,
    maxTokens: 6000,
  });

  return { final, specialists };
}

Three readable files — config.js (your roster) · orchestrator.js (the parallel core) · bot-discord.js (the !team command). No framework, no config wall. Yours to fork.

Who it's for

Built for people who ship.

Runflock is opinionated. It's a great fit for some, a bad fit for others — here's the honest split.

✓ Built for

Devs who already use Claude daily and want a repeatable multi-agent workflow
Engineers who want to read every line — no black box
Teams who fire !team at problems from Discord
Anyone shipping who'd rather own the code than rent a SaaS

✕ Not for

No-code seekers who want a hosted dashboard to log into
Non-developers — this runs from your terminal
Anyone who won't bring their own Anthropic API key
People who want someone else to run it as a subscription

Pricing

One payment. Yours forever.

No subscription, no seats, no per-run fee. Just the code — and every future update.

The Starter Kit

A week of orchestration plumbing · ~$3,000 of your time

$129 · one-time

You keep the week. And the code — forever.

Full source: orchestrator, agents & Discord bot — yours to fork
Parallel multi-agent engine on the official Anthropic SDK
Streaming + adaptive thinking, wired up correctly
Fully customizable roster — add, rename, clone any agent
Running in under 5 min · commercial license · unlimited projects
Lifetime updates — every future improvement, free

Get the kit — $129

🔒 Secure checkout via Stripe · Instant download link on-screen + emailed · you own the code

It runs, or I help you personally. Hit a snag? Reply to your receipt and I'll get it working with you — a real human, no hoops.

FAQ

The honest answers.

Isn't this just a wrapper around the API?

Yes — and that's the point. It's a thin, readable wrapper that does the hard part: running agents in parallel and synthesizing their output into one answer, with streaming and adaptive thinking done right. You're paying to skip the fiddly week of getting orchestration correct, not for a black box.

What are the API costs?

You bring your own Anthropic API key and pay Anthropic directly — the kit adds zero markup. A typical 3-agent run is a handful of cents to low dollars depending on goal size. You control the roster and models, so you control the bill.

Do I need a subscription?

No. $129 is a one-time payment for the code. The only ongoing cost is your own Anthropic API usage, billed by Anthropic. No SaaS, no seats, no per-run fee to me.

Can I actually customize the agents?

Completely. Each agent is a plain object — name, role, system instructions. Run two Researchers and a Security Critic, change how many fire in parallel. It's your code; there's no config wall between you and the behavior.

Which models does it use?

It ships pointed at Claude Opus 4.8 via the official Anthropic SDK. Want a cheaper model for the specialists and Opus only for the lead synthesis? One-line change per agent.

What if I can't get it running?

Reply to your receipt email and I'll personally help you get it running — a real human, not a bot, until it works. Because it's downloadable source you own the second you buy, there are no refunds — instead I make sure you actually win with it. No hoops.

Stop prompting one AI.
Run a whole team.