shellshare CLAUDE.md

Project instructions for Shellshare, a Rust program that broadcasts a live terminal session through web links. The same program can act as a client, server, or combined local service.

In plain words
What is it for?
Building and linting Shellshare, running its server or client, sharing a local terminal, and executing browser-based end-to-end tests.
Why use it?
They explain the repository's build, lint, run, and end-to-end test process. This helps agents work safely with its dual-mode architecture and Rust-specific checks.

Instructions file

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/vitorbaptista/shellshare/claude-md
Clone the repo
git clone --depth 1 https://github.com/vitorbaptista/shellshare
Per session 3,612 This file is loaded in full into every session.
When invoked 3,612 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 $0.03612 $0.03612
Opus 5 $0.01806 $0.01806
Sonnet 5 $0.00722 $0.00722
Haiku 4.5 $0.00361 $0.00361

Measured 3d ago against content hash cf8de759540e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

shellshare CLAUDE.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 3d 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.

CLAUDE.md · 127 lines

How it starts

The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

Shellshare is a live terminal broadcasting tool that allows users to share their terminal session via web links. The project is a complete Rust rewrite providing both client and server in a single binary.

Build & Development Commands

# Build
make build                     # Same as above

# Lint (pedantic clippy lints are enabled)
make lint                      # Both clippy + check

# Run locally
cargo run -- server            # Start server on 0.0.0.0:3000
cargo run -- server --port 8080 --host 127.0.0.1
cargo run -- --server http://localhost:3000  # Run client
cargo run -- serve             # Share via a local server on localhost:3000

# E2E tests (Python + Playwright)
# Requires a release binary: the suite starts its own servers from it
cargo build --release
cd e2e && uv sync && uv run pytest -n 10

Architecture

Dual-mode binary: shellshare operates as client (default) or server (shellshare server). shellshare serve combines both: it boots the embedded server on a background thread (default localhost:3000, configurable via --host/--port) and runs the client against it, sharing the terminal with no external server. Both serve and server accept --tunnel (src/tunnel.rs): it spawns the user's pre-installed cloudflared against the local server, waits for the https://*.trycloudflare.com URL from its stderr banner, and uses it as the share link (the broadcaster still talks to localhost); missing cloudflared is a fatal error pointing at the install docs, and the tunnel process dies with shellshare.

Scripting surface: shellshare exec -- <cmd> runs one command in the PTY (instead of a shell), broadcasts it, and exits with the command's exit code. The global --json flag switches stdout to newline-delimited JSON events: first one with event sharing (parse its url field), last {"event":"end","exit_code":N} (errors stay on stderr as ERROR: ...). This contract is documented in AGENTS.md and templates/llms.txt and covered by e2e/test_agents.py - the three must stay in lockstep.

Read the full file on GitHub · 127 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. 3d ago First seen · 127 lines · 3,612 tokens per session scan A cf8de759540e

Subscribe to this mod's changes

shellshare CLAUDE.md is an instructions file published in the GitHub repository vitorbaptista/shellshare (226 stars, last pushed 9d ago), licensed Apache-2.0. It adds 3,612 tokens to every session, about $0.0181 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.