OpenSwarm AGENTS.md

Customization instructions for OpenSwarm, a multi-agent AI project where specialist agents work together through a shared coordinator.

In plain words
What is it for?
Use them when adding or customizing specialist agents, their tools, shared instructions, orchestration, terminal runs, or the API server.
Why use it?
They explain how agents, shared instructions, tools, and entry points fit together, making it easier to change the team without breaking its connections.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/vrsen/openswarm/agents-md
Clone the repo
git clone --depth 1 https://github.com/VRSEN/OpenSwarm

Made for: Codex, OpenCode.

Per session 1,053 This file is loaded in full into every session.
When invoked 1,053 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01053 $0.01053
Opus 5 $0.00526 $0.00526
Sonnet 5 $0.00211 $0.00211
Haiku 4.5 $0.00105 $0.00105

Measured 2d ago against content hash 71a3168c70d6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

OpenSwarm AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 131 lines

How it starts

The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OpenSwarm — Customization Guide

This file gives coding agents (Cursor, Claude Code, Codex, etc.) everything they need to understand and customize this swarm. Read it before making any changes.


What is OpenSwarm?

OpenSwarm is a multi-agent AI team you can fork and reshape into any kind of swarm you need — SEO, sales, research, finance, customer support, or anything else. Each agent is a specialist. They collaborate through a shared orchestrator.


Folder Structure

swarm.py                  ← main config: imports all agents, defines how they connect
shared_instructions.md    ← context shared across every agent
run.py                    ← CLI entry point (terminal demo)
server.py                 ← API entry point (FastAPI server)

orchestrator/
  orchestrator.py         ← agent definition
  instructions.md         ← system prompt

data_analyst_agent/
  data_analyst_agent.py
  instructions.md
  tools/                  ← custom tools for this agent

docs_agent/
  docs_agent.py
  instructions.md
  tools/

slides_agent/
  slides_agent.py
  instructions.md
  tools/

image_generation_agent/
  image_generation_agent.py
  instructions.md
  tools/

video_generation_agent/
  video_generation_agent.py
  instructions.md
  tools/

deep_research/
  deep_research.py
  instructions.md
  tools/

virtual_assistant/
  virtual_assistant.py
  instructions.md
  tools/

shared_tools/             ← tools available to all agents (Composio integrations, etc.)

How Agents Connect (swarm.py)

swarm.py is the only file you need to edit when adding, removing, or rewiring agents. It:

  1. Imports a create_* factory function from each agent folder
  2. Instantiates all agents
  3. Defines communication flows — who can talk to whom

The default pattern is orchestrator-to-all: the orchestrator can send messages to every specialist, and all agents can hand off to each other.


How to Customize

To build your own swarm from this repo:

  1. Fork and rename the repo (e.g., seo-swarm)
  2. Decide which agents to keep, rename, or replace
    • Rename the folder and its files to match the new agent's purpose
    • Update instructions.md with the new system prompt
    • Update swarm.py to import and register the renamed agent
  3. Add or remove tools inside each agent's tools/ folder
  4. Update shared_instructions.md with any context all agents should share
  5. Run with python run.py

Read the full file on GitHub · 131 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 131 lines · 1,053 tokens per session scan A 71a3168c70d6

Subscribe to this mod's changes

OpenSwarm AGENTS.md is an instructions file published in the GitHub repository VRSEN/OpenSwarm (2,855 stars, last pushed 1mo ago), licensed MIT. It adds 1,053 tokens to every session, about $0.0053 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.