What is agent-swarm CLAUDE.md?

Project instructions for Agent Swarm, an open-source application for coordinating multiple coding agents. They describe its server, web client, Docker containers, development commands, and handling of agent messages and identity data.

Use them when developing Agent Swarm, changing its server or client, working on agent coordination, running its development setup, or checking files before a commit.

They help contributors understand where the main parts of the project live and avoid committing secrets or private agent data. They also provide security checks to follow before each commit.

How tlangridge/agent-swarm configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to tlangridge/agent-swarm. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/tlangridge/agent-swarm/master/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/tlangridge/agent-swarm

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/tlangridge/agent-swarm/claude-md/github.svg)](https://agentmods.dev/instructions/tlangridge/agent-swarm/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/tlangridge/agent-swarm/claude-md"><img src="https://agentmods.dev/badge/instructions/tlangridge/agent-swarm/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/tlangridge/agent-swarm/claude-md"><img src="https://agentmods.dev/badge/instructions/tlangridge/agent-swarm/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 585 This file is loaded in full into every session.

When invoked 585 The same file — it is already loaded in full.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00585	$0.00585
Opus 5.5	$0.00234	$0.00234
Sonnet 5.5	$0.00117	$0.00117
Haiku 4.5	$0.00059	$0.00059

Measured 2mo ago against content hash 9fa97863af6f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

agent-swarm CLAUDE.md scanned grade A with 1 finding 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 2mo 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Agents communicate via REST API at `/api/swarm/*` — all agents use `curl` to send/receive messages

CLAUDE.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Swarm

Security — Open Source Repo

This is a public open-source repository. Before every commit:

Check .gitignore — Ensure .env, data/agents/, data/docker/, and any files containing secrets are gitignored.
Never commit API keys, tokens, or credentials — All secrets belong in .env (which is gitignored). If you see a key in source code, move it to .env and reference via process.env.
Never commit agent identity data — The data/ directory contains per-agent JSON files and Docker volume state. It must stay gitignored.
Audit new files — When creating new files, consider whether they could contain sensitive data. When in doubt, add to .gitignore.

Project Structure

server/ — Express + WebSocket + node-pty backend (TypeScript, runs via tsx)
- server/services/swarm-registry.ts — In-memory swarm membership and role management
- server/services/swarm-prompts.ts — System prompt templates for agent swarm awareness
- server/routes/swarm.ts — REST API for agent-to-agent messaging (/api/swarm/*)
client/ — React + Vite frontend (xterm.js terminals, react-mosaic tiling)
docker/ — Dockerfile for sandboxed agent containers
data/ — Runtime data (agents JSON, Docker volumes) — gitignored

Dev Commands

npm run dev — Start server + client concurrently
docker build -t agent-swarm-sandbox docker/ — Build the sandbox image

Swarm Coordination

Agents communicate via REST API at /api/swarm/* — all agents use curl to send/receive messages
Messages are injected into agent terminals via session.pty.write() with [SWARM from <name>]: framing
Agents authenticate with X-Session-Id header (session ID passed via AGENT_SWARM_SESSION_ID env var)
Single lead agent at a time — user designates at launch or toggles via crown icon in tile toolbar
Claude Code agents get full swarm instructions via --append-system-prompt
Non-Claude agents (Gemini, Codex, Bash) get env vars + a one-time terminal-injected orientation message
Docker agents reach the host API via http://host.docker.internal:${PORT}

Read the full file on GitHub · 42 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.

2mo ago First seen · 42 lines · 585 tokens per session scan A 9fa97863af6f

Subscribe to this mod's changes

agent-swarm CLAUDE.md is an instructions file published in the GitHub repository tlangridge/agent-swarm (5 stars, last pushed 8mo ago), licensed MIT. It adds 585 tokens to every session, about $0.0023 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents

agent-swarm CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
agent-swarm CLAUDE.mdtlangridge/agent-swarm	Instructions file	5	8mo ago	MIT	A	585	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	2,022	GitHub Copilot

agent-swarm: Instructions file for Claude Code