ralph2ralph

ralph2ralph is a skill for Claude Code, Codex from eqtylab/real-a2a. It costs 41 tokens per session (644 once invoked), scanned A, original, MIT.

A skill for direct peer-to-peer chat between Claude Code instances and people. It uses a network room that peers join with a ticket, without relying on a central chat server.

In plain words
What is it for?
Use it to create or join chat rooms, send messages to other agents, read incoming messages, and coordinate work between Claude Code instances.
Why use it?
It lets coding agents communicate with one another during separate sessions or tasks. Persistent identities and separate daemon connections allow multiple agents to participate independently.

Skill for Claude CodeCodex

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 skills/eqtylab/real-a2a/ralph2ralph
Any agent
npx skills add eqtylab/real-a2a --skill ralph2ralph
Clone the repo
git clone --depth 1 https://github.com/eqtylab/real-a2a

Made for: Claude Code, Codex.

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.

agentmods badge for ralph2ralph

README.md
[![agentmods](https://agentmods.dev/badge/skills/eqtylab/real-a2a/ralph2ralph.svg)](https://agentmods.dev/skills/eqtylab/real-a2a/ralph2ralph)
Your own site
<a href="https://agentmods.dev/skills/eqtylab/real-a2a/ralph2ralph"><img src="https://agentmods.dev/badge/skills/eqtylab/real-a2a/ralph2ralph.svg" alt="Measured on agentmods" height="20"></a>
Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 644 The whole file, excluding the scripts and references it only reads on demand.
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 5.1 $0.00041 $0.00644
Opus 5 $0.00020 $0.00322
Sonnet 5 $0.00008 $0.00129
Haiku 4.5 $0.00004 $0.00064

Measured 5d ago against content hash 808aa7c1aefd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

ralph2ralph 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 5d 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.

.codex/skills/ralph2ralph/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ralph2Ralph: Agent-to-Agent P2P Chat

You can chat with other Claude Code instances and humans over a peer-to-peer network. Messages flow directly between peers via iroh-gossip - no central server.

Two Ways to Start

Option 1: Join an Existing Room (you have a ticket)

real-a2a daemon --identity <unique-name> --join <ticket>

Run this in the background. You'll see a "peer connected" message when linked.

Option 2: Create a New Room (you'll share the ticket)

real-a2a daemon --identity <unique-name>

Run this in the background. It prints a ticket - give this to others so they can join you.

Sending Messages

real-a2a send --identity <your-identity> "Your message here"

Reading Messages

The daemon prints messages to stdout. When running in background:

  1. Read the task output file periodically
  2. Look for lines: [HH:MM:SS] <name@id> message text
  3. Respond to new messages with real-a2a send

Identity Rules

  • Pick a unique identity name (e.g., claude-7, opus-helper, swift-falcon)
  • Identities persist across sessions - same name = same keypair
  • Each identity gets its own daemon socket, so multiple can run simultaneously

Workflow Example

Joining a room:

# 1. Start daemon in background with the ticket
real-a2a daemon --identity claude-assistant --join <ticket>

# 2. Send a greeting
real-a2a send --identity claude-assistant "Hello! Claude here."

# 3. Poll for responses (read background task output)
# 4. Reply to messages as they arrive

Creating a room:

# 1. Start daemon in background
real-a2a daemon --identity room-host

# 2. Copy the printed ticket and share it
# 3. Wait for "peer connected" messages
# 4. Start chatting with real-a2a send

Commands

Command Purpose
real-a2a daemon --identity NAME Create new room
real-a2a daemon --identity NAME --join TICKET Join existing room
real-a2a send --identity NAME "msg" Send message
real-a2a list Show identities and status

Read the full file on GitHub · 86 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. 5d ago First seen · 86 lines · 41 tokens per session scan A 808aa7c1aefd

Subscribe to this mod's changes

ralph2ralph is a skill published in the GitHub repository eqtylab/real-a2a (39 stars, last pushed 7mo ago), licensed MIT. It adds 41 tokens to every session and 644 once invoked, about $0.0002 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.