launch-pad CLAUDE.md

Repository instructions for Launch Pad, a self-hosted deployment tool. It builds an application into a Docker image, sets up selected AWS services, runs the app on the user's own server and enables HTTPS through Caddy.

In plain words
What is it for?
Working on the deployment command-line tool, Docker builds, AWS setup, server configuration, HTTPS and automatic updates.
Why use it?
They explain the deployment architecture and where to find the relevant documentation, reducing errors in commands, configuration and cloud provisioning.

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/agentsystemlabs/launch-pad/claude-md
Clone the repo
git clone --depth 1 https://github.com/AgentSystemLabs/launch-pad
Per session 7,830 This file is loaded in full into every session.
When invoked 7,830 The same file — it is already loaded in full.
Security scan A 1 finding. 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 $0.07830 $0.07830
Opus 5 $0.03915 $0.03915
Sonnet 5 $0.01566 $0.01566
Haiku 4.5 $0.00783 $0.00783

Measured 2d ago against content hash 6af728001fcc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

launch-pad 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 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.

Makes network callslowCapability

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

role-specific binary is uploaded to `nodes/<id>/agent` in S3 and the node `curl`s it via a
CLAUDE.md · 407 lines

How it starts

The opening of the file, as written. The whole thing — 407 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.

What this is

Launch Pad is a self-hosted deploy tool: one command (launchpad deploy) builds a user's app into a Docker image, pushes it to ECR, provisions AWS (EC2 + IAM + S3), runs it on the user's own node behind Caddy with automatic HTTPS, and auto-updates on the next deploy. The north-star spec is docs/overview.md — read it before non-trivial work.

Start with README.md — it is the documentation directory for the project: intent, positioning, and a table of per-area deep dives in docs/*.md. Before working in an unfamiliar part of the repo, read the matching doc: docs/codebase-layout.md (repo map + "where to change what"), docs/architecture.md, docs/cli.md, docs/configuration.md, docs/agent.md, docs/golden-ami.md, docs/dashboard.md, docs/testing.md. When a change alters user-facing behavior (commands, config fields, provisioning, wire contracts), update the matching docs/*.md and, if the surface list changes, the README table.

The whole system is declarative, with no control-plane server: the CLI writes desired state to S3; an agent on each node polls S3 and reconciles Docker + Caddy to match. S3 is the only thing the two sides share.

CLI (local) ──writes desired.json──▶ S3 ◀──polls desired.json── agent (on node)
CLI (local) ◀──polls status.json──── S3 ──writes status.json──▶ agent (on node)

Commands

pnpm workspace (Node ≥ 24, pnpm 11). Run from the repo root unless noted.

pnpm build            # build all packages (tsup) — see ordering note below
pnpm typecheck        # tsc --noEmit across all packages
pnpm test             # vitest run across all packages
pnpm dev              # all packages in watch/run mode in parallel
pnpm clean            # remove dist / tsbuildinfo

# One package
pnpm --filter @agentsystemlabs/launch-pad-shared test
pnpm --filter @agentsystemlabs/launch-pad        typecheck

# One test file / one test name (extra args forward to vitest)
pnpm --filter @agentsystemlabs/launch-pad-shared test src/config.test.ts
pnpm --filter @agentsystemlabs/launch-pad-shared test -- -t "capacity"

# Run the CLI locally without building (workspace dev)
pnpm --filter @agentsystemlabs/launch-pad dev -- deploy --dry-run

# Link the CLI globally from another directory (no build — runs TypeScript via tsx)
npm link                              # from repo root; bins invoke packages/cli/src on every run
# Test the compiled artifact instead: pnpm link:dist

Read the full file on GitHub · 407 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 · 407 lines · 7,830 tokens per session scan A 6af728001fcc

Subscribe to this mod's changes

launch-pad CLAUDE.md is an instructions file published in the GitHub repository AgentSystemLabs/launch-pad (22 stars, last pushed 1mo ago), licensed MIT. It adds 7,830 tokens to every session, about $0.0392 per session on Opus 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 2026-08-30.