setup-forge

setup-forge is a command for Claude Code from ForgeyClap/claude-forge. It costs 53 tokens per session (2,890 once invoked), scanned B, original, MIT.

A first-run setup wizard for Forge, a project system that helps coding agents plan and carry out work. It asks about the project, optionally configures API keys, and creates the project’s Forge files.

In plain words
What is it for?
Use it once when starting Forge in a project, or use its checks and reset options when managing the setup.
Why use it?
It gives new users a guided setup and helps keep secrets out of files tracked by version control.

Command for Claude Code

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 commands/forgeyclap/claude-forge/setup-forge
Clone the repo
git clone --depth 1 https://github.com/ForgeyClap/claude-forge

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.

agentmods badge for setup-forge

README.md
[![agentmods](https://agentmods.dev/badge/commands/forgeyclap/claude-forge/setup-forge.svg)](https://agentmods.dev/commands/forgeyclap/claude-forge/setup-forge)
Your own site
<a href="https://agentmods.dev/commands/forgeyclap/claude-forge/setup-forge"><img src="https://agentmods.dev/badge/commands/forgeyclap/claude-forge/setup-forge.svg" alt="Measured on agentmods" height="20"></a>
Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,890 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 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 $0.00053 $0.02890
Opus 5 $0.00026 $0.01445
Sonnet 5 $0.00011 $0.00578
Haiku 4.5 $0.00005 $0.00289

Measured yesterday against content hash 767e241e99a4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

setup-forge scanned grade B 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 yesterday.

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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

values into the gitignored `.env`, best-effort `chmod 600`, and writes a values-free
global-install/.claude/commands/setup-forge.md · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/setup-forge — Forge first-run onboarding wizard (global core)

Run the Forge onboarding for: $ARGUMENTS

This is the bare /setup-forge installed into ~/.claude by the Forge installer. It runs in the current project folder. You (the Lead agent) drive a friendly, beginner-safe onboarding; the mechanical and security-sensitive steps are done by a zero-dependency engine you call — node .claude/forge-bin/forge-setup.cjs (resolved in the current project) — never by hand.

Non-negotiable honesty & safety rules

  • Never print or echo a secret value — names only, never "to confirm".
  • Never claim a step ran if it didn't. Report exactly what the engine returned.
  • Never write a secret to a git-tracked file. The engine enforces this; don't shortcut it.
  • Never delete or overwrite the user's own files without saying so first.
  • Keys are optional — a newcomer can skip and still finish. Forge runs without keys.
  • Work only in this project folder. Don't touch other projects. The only global writes are the documented markers.

Global route — locate the engine first

Because this command lives in ~/.claude, the project may or may not have the full Forge payload yet. Check before anything else:

  • If .claude/forge-bin/forge-setup.cjs exists in the current project → use it (normal flow below).
  • If it is missing, the per-project payload isn't installed here. Say so plainly and offer the fix: run the installer (install.sh on macOS/Linux, install.ps1 on Windows) from the claude-forge repo, or copy .claude/ into this project manually. Once present, re-run /setup-forge. Do not fabricate a key flow or dashboard that isn't installed.

Language first (internationalization)

Determine the working language so the entire wizard and every confirmation run in it:

  1. node .claude/forge-bin/forge-setup.cjs lang (prints en default) — if the engine is absent, fall back to English and to any --lang flag.
  2. A --lang <code> flag in $ARGUMENTS overrides.
  3. On a fresh run, ask it as question 4 below.

Read the full file on GitHub · 174 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. yesterday First seen · 174 lines · 0 tokens per session scan B 767e241e99a4

Subscribe to this mod's changes

setup-forge is a command published in the GitHub repository ForgeyClap/claude-forge (2 stars, last pushed 1mo ago), licensed MIT. It adds 53 tokens to every session and 2,890 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.