swarma

Coordinate agent teams to run the GROWS loop for growth experiments.

175|14|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/glitch-rabin/swarma --skill swarma
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: swarma
Source: https://github.com/glitch-rabin/swarma/tree/main
Command: npx skills add https://github.com/glitch-rabin/swarma --skill swarma

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

swarma provides a turnkey framework to run repeated growth experiments with autonomous agent teams, eliminating manual guesswork and unstructured exploration.

Core Features & Use Cases

  • GROWS loop: generate hypotheses, run experiments, observe signals, weigh verdicts, and stack playbooks to improve strategy over time.
  • 18 pre-built squads: ready-to-run templates covering the full AARRR funnel for fast start.
  • Auto-evolution: strategies and playbooks evolve automatically from results to build a living knowledge base.

Quick Start

Run a starter cycle with the built-in starter team to verify the workflow and costs.

Frequently Asked Questions about swarma

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate growth experiments across the full AARRR funnel?

You can automate growth experiments by deploying cross-functional agent teams that execute the GROWS loop to generate hypotheses, run tests, observe signals, and stack playbooks. It provides 18 pre-built squads covering the full AARRR funnel for immediate scaling.

What is the GROWS loop framework for optimizing marketing strategies?

The GROWS loop is an iterative process where autonomous agents generate hypotheses, run experiments, observe signals, weigh verdicts, and stack playbooks. This mechanism allows your growth strategies to auto-evolve based on experimental results, building a living knowledge base.

Do I need a local LLM instance and MCP integration to run autonomous growth agents?

Yes, running autonomous growth agents requires a local instance equipped with an OpenRouter-compatible LLM, a knowledge store, and MCP/Hermes integration. These components provide the necessary orchestration and environment to coordinate cross-functional agent teams.

How do I start running A/B testing playbooks with agent squads?

You start by running a starter cycle with the built-in starter team to verify the workflow and costs. Once verified, you can scale up to deploy the 18 pre-built squads to automate A/B testing and growth experiments across your marketing funnel.

What is the best way to scale marketing experiments without manual guesswork?

The best way to scale marketing experiments without manual guesswork is using a turnkey framework of autonomous agent teams. These agents coordinate to run the GROWS loop continuously, evolving your strategies automatically from observed results to eliminate unstructured exploration.