reforge-orchestrator

reforge-orchestrator is an agent for Claude Code from ByeongminLee/nextjs-claude-code. It costs 79 tokens per session (2,577 once invoked), scanned C, original, MIT.

A workflow that studies an existing codebase and turns it, together with requested changes, into structured specification documents. It reads the legacy project without changing it and writes the specifications into the current project.

In plain words
What is it for?
Use it to analyse legacy applications, answer questions about how they work, compare requested changes with the existing system, and prepare implementation specifications.
Why use it?
It provides a written understanding of old software and the proposed changes before anyone starts rewriting code.

Agent for Claude Code

Part of the nextjs-claude-code plugin — 9 skills, 49 agents shipped together

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 agents/byeongminlee/nextjs-claude-code/reforge-orchestrator
Clone the repo
git clone --depth 1 https://github.com/ByeongminLee/nextjs-claude-code

Made for: Claude Code.

Or install nextjs-claude-code, the plugin that ships this one along with the rest of its 9 skills, 49 agents.

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 reforge-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/byeongminlee/nextjs-claude-code/reforge-orchestrator.svg)](https://agentmods.dev/agents/byeongminlee/nextjs-claude-code/reforge-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/byeongminlee/nextjs-claude-code/reforge-orchestrator"><img src="https://agentmods.dev/badge/agents/byeongminlee/nextjs-claude-code/reforge-orchestrator.svg" alt="Measured on agentmods" height="20"></a>
Per session 79 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,577 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 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 $0.00079 $0.02577
Opus 5 $0.00039 $0.01288
Sonnet 5 $0.00016 $0.00515
Haiku 4.5 $0.00008 $0.00258

Measured 4d ago against content hash d4c78376c9f4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

reforge-orchestrator scanned grade C 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 4d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

"Reforge complete. All information has been captured in the spec documents. You can safely remove the legacy folder: `rm -rf [LEGACY_PATH]`"
template/.claude/agents/reforge-orchestrator.md · 286 lines

How it starts

The opening of the file, as written. The whole thing — 286 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a legacy-to-spec transformation orchestrator. You guide users from an existing codebase to a complete set of NCC-format spec documents, incorporating requested changes.

You do NOT write source code. You do NOT modify the legacy folder. You write to spec/reforge/[name]/ and spec/feature/[name]/.

Directory Model

project-root/              ← Current NCC project (write target)
├── .claude/
├── spec/                  ← Specs generated here
│   ├── reforge/[name]/    ← Reforge working documents
│   └── feature/[name]/    ← Standard NCC feature specs
└── [legacy-path]/         ← Legacy project (read-only source)

Resume Protocol

Check spec/reforge/*/SOURCE.md for Status: analyzing:

Artifact exists Resume from
None / no SOURCE.md Phase 0 (fresh start)
ANALYSIS.md Phase 2
CHANGES.md Phase 3
DELTA.md Phase 4
DECISION.md Status: completed Already done — inform user

Inform user: "Resuming reforge [name] from Phase [N]."

Phase 0: Setup

  1. Detect user's language from $ARGUMENTS (Korean, English, Japanese, etc.). All subsequent questions and outputs must be in this language.

  2. Verify legacy folder: Confirm LEGACY_PATH exists and contains package.json.

    • If not → report error and STOP.
  3. Derive reforge name from the legacy project's package.json name field or directory name (kebab-case, English). If ambiguous, ask user to confirm.

  4. Create working directory: spec/reforge/[name]/

  5. Record source: Write spec/reforge/[name]/SOURCE.md:

    # Reforge Source
    Legacy path: [LEGACY_PATH]
    Started: YYYY-MM-DD
    Status: analyzing
    Phase: 0
    
  6. Update STATE.md: Add a reforge tracking entry:

    ### [reforge-name] [reforging]
    Started: YYYY-MM-DD
    Source: [LEGACY_PATH]
    Phase: 0/5
    

Phase 1: Deep Analysis

Spawn codebase-analyzer agent (sonnet) to deeply analyze the legacy project:

[HANDOFF]
TO: codebase-analyzer (sonnet)
TASK: Deep analysis of legacy codebase at [LEGACY_PATH]
REFORGE_NAME: [name]
DONE-WHEN: ANALYSIS.md written to spec/reforge/[name]/
READS: [LEGACY_PATH]/** (read-only)
OUTPUT: spec/reforge/[name]/ANALYSIS.md
[/HANDOFF]

Read the full file on GitHub · 286 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. 4d ago First seen · 286 lines · 79 tokens per session scan C d4c78376c9f4

Subscribe to this mod's changes

reforge-orchestrator is an agent published in the GitHub repository ByeongminLee/nextjs-claude-code (3 stars, last pushed 5mo ago), licensed MIT. It adds 79 tokens to every session and 2,577 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

stitch-kit

Use this agent for anything Stitch-related: ideating designs through conversation, generating UI screens from text, editing/iterating designs, generating design variants, managing Stitch Design Systems, converting designs to production code, extracting design tokens, and running the full design-to-ship pipeline.…

gabelul/stitch-kit · 237 tokens

fe-a11y-auditor

Specialized accessibility (a11y) audit — semantic HTML, ARIA, keyboard navigation, color contrast, focus management. More precise than fe-reviewer's a11y axis. READ-ONLY. With --live, also measures via Chrome DevTools MCP (Lighthouse a11y score, real accessibility tree, runtime console); falls back to static analysis…

sh5623/fe-rail · 85 tokens

fe-perf-auditor

In-depth React performance audit — bundle size, data fetching, image optimization, dynamic import, Suspense. Supports both Next.js (RSC, next/image) and Vite SPA (bundle analysis, fetchpriority). More precise than fe-reviewer's performance axis. READ-ONLY. With --live, also measures via Chrome DevTools MCP…

sh5623/fe-rail · 92 tokens

fe-git-operator

Dedicated git operations — splitting commits, safe staging, and writing Conventional Commits bodies (fix = symptom/cause/fix, feat = addition/core/impact). fe-pr-author owns PRs; this agent owns commits. Destructive commands forbidden.

sh5623/fe-rail · 55 tokens

fe-reviewer

4-axis frontend code review — type safety, performance, a11y, code quality. READ-ONLY, no direct edits. Delegate after fe-build completes, before the PR stage.

sh5623/fe-rail · 41 tokens

fe-vision

Precisely extracts layout, components, color, and typography from a single screen in Figma, screenshots, UI mockups, design drafts, or PDF/PPT (via conversion) — extraction mode. Or compares an implementation screenshot against a reference and judges visual fidelity as JSON — comparison mode (visual-verdict).…

sh5623/fe-rail · 90 tokens