Free dev tools that track your AI coding spend.

RoninForge is one independent developer's forge: free, open-source dev tools that track AI coding spend. A local spend monitor for Claude Code, a web dashboard for team AI spend, and an open index of dated model prices, so you can track what AI coding assistants really cost. Local-first, no accounts, MIT licensed.

Browse tools Browse plugins
A ronin forging code at an anvil

Free Data Collections

Open, first-party datasets, free to browse, cite, and download. CC BY 4.0.

AI Price Index

Chart how AI model API token prices changed over time. Filter by model and variation, read the price on any past date, and export the chart. Built on open, first-party-sourced data, CC BY 4.0.

Explore prices

State of MCP

A zero-key census of all 19,804 servers in the official Model Context Protocol registry, classified healthy, degraded, or dead, with a page for each one. Also scores every reachable server against the 2026-07-28 spec revision. Dated editions, CC BY 4.0.

Read the latest report

Free Web Tools

No install, no account, no upload. Everything runs in your browser.

LLM API pricing comparison

Claude, GPT, Gemini, DeepSeek, Mistral, and Grok token prices side by side, verified against the official pages.

Compare

GitHub AI Credits calculator

Estimate your monthly credit burn under GitHub Copilot's usage-based billing and see whether your plan allowance covers it.

Calculate

GitHub AI Credits explained

How Copilot's usage-based billing works: per-model credit rates, plan allowances, overage, and pooling, with a page for every model and plan.

Read the reference

Codex credits calculator

OpenAI Codex meters usage in credits worth $0.04 each. Estimate GPT-5.6 Sol, Terra, and Luna credit burn per workload and see that it equals the direct API rate.

Calculate

Codex credits explained

What a Codex credit is worth, why credits are OpenAI API pricing repackaged with no markup, and a page for every model, amount, and comparison.

Read the reference

Copilot bill migrator

Moved from premium requests to AI credits on June 1? Compare your old and new monthly bill and see if usage-based billing costs you more.

Compare bills

Copilot usage CSV analyzer

Drop the usage report GitHub emails you and see credits by model, day, SKU, and user. The file never leaves your browser.

Analyze

Claude Code spend analyzer

Drop your ~/.claude/projects folder and see your Claude Code usage value by project, git branch, and model. Parsed in your browser, nothing uploaded.

Analyze

AI stack cost auditor

Tick your AI subscriptions and add your API spend to get one honest monthly total with a breakdown. Subscriptions plus usage, in one number. Nothing you enter leaves your browser.

Total your spend

Anthropic API cost analyzer

Drop your Anthropic usage export and see Claude API cost by API key, model, and day. Parsed in your browser, nothing uploaded.

Analyze

Tools

BudgetClaw

v1.7.7

Local spend monitor for Claude Code. Tracks cost per project and git branch, warns you before a cap, and kills the process when you blow past it.

  • Per-project, per-branch budget caps (daily / weekly / monthly)
  • Phone alerts via ntfy -- install the app, subscribe to a topic, run budgetclaw alerts setup
  • SIGTERM on breach + lockfile that catches silent relaunches
  • Point-in-time pricing from the open AI Price Index dataset (CC BY 4.0), embedded at build time so it works offline by default and prices each event at the rate in effect when it ran, so a later price change never restates your recorded history
  • Optional budgetclaw sync pushes aggregate spend rollups to a Goei dashboard. Only dollar and token totals leave the machine, never prompts or keys.
  • Zero keys. Zero prompts. Reads local logs only.
  • Single Go binary, 5 MB, no runtime, no account
About BudgetClaw
$ budgetclaw status

PROJECT     BRANCH        TODAY    WEEK
myapp       main          $4.85    $28.30
myapp       feature/auth  $1.20    $3.20
side-proj   main          $0.00    $0.50
TOTAL                     $6.05    $32.00

$ budgetclaw limit list

#  PROJECT  PERIOD  CAP      ACTION
1  *        daily   $10.00   warn
2  myapp    daily   $5.00    kill
Goei beta

Your AI cost bodyguard. A web dashboard for where your AI money goes, per project, branch, and model. A zero-key local track reads what Claude Code already writes on your machine; an optional billed track adds Anthropic and ElevenLabs.

  • Where your Claude Code money goes, per project, branch, and model
  • Zero-key local track, fed by the open-source budgetclaw CLI
  • Billed vs usage value for Pro and Max subscribers
  • Budget alerts via phone push, webhooks, and RSS
Open dashboard How the spend dashboard works
Goei AI Cost Dashboard

Anthropic

Claude Opus 4, Sonnet 4

$412.80

this month

ElevenLabs

3,205 / 10,000 characters

32%

of quota

More providers coming

OpenAI, Stability, ...

> /claude-code-cost:report

PROJECT     BRANCH        TODAY    WEEK     MONTH
myapp       main          $4.85    $28.30   $112.40
myapp       feature/auth  $1.20    $3.20    $3.20
side-proj   main          $0.00    $0.50    $14.10
TOTAL                     $6.05    $32.00   $129.70

claude-code-cost

v1.1.0 Claude Code plugin

The zero-install way to see your Claude Code spend. Install it inside Claude Code, then run /claude-code-cost:report for spend per project and git branch, or :by-day / :by-week / :by-month over time. No binary, no account.

  • Installs inside Claude Code through the plugin system. No binary to download.
  • /claude-code-cost:report shows spend per project and git branch; :by-day, :by-week, :by-month, and :by-project break it down over time. Every command takes an optional substring filter
  • Point-in-time pricing from the open AI Price Index dataset, bundled so it works offline by default
  • Zero keys. Zero prompts. Reads local session logs only.
  • Uses BudgetClaw automatically when it is installed, for budget caps and a persistent history
About claude-code-cost

Hanko

v0.2.1

Validate Claude Code plugin manifests before submission. Catches reserved marketplace names, duplicate hooks declarations, path traversal, and the schema errors the official validator reports opaquely. Go binary, GitHub Action, zero network calls.

About Hanko

判子 · the personal seal stamped on official Japanese documents.

Tsuba

v0.2.1

Scaffold marketplace-ready Claude Code skills and plugins in seconds. Every generated plugin ships with a correct plugin.json, a sample skill, LICENSE, and README, and passes hanko validation on the first run.

About Tsuba

鍔 · the katana guard that lets a swordsman hand the blade off safely.

Akashi

v0.3.0

Verify a Model Context Protocol server is alive and conformant, keyless. Probes the registry, GitHub, npm, PyPI, and a capability-only MCP initialize handshake, then reports healthy, degraded, or dead and hands you an embeddable verified badge.

About Akashi State of MCP data

証 · the proof or certificate that something is genuine.

Cursor Plugins

Thirteen framework plugins for Cursor: verified rule packs, migration skills, and anti-pattern detection for Tailwind v4, Django, Vue and more. They are stable and receive fixes only; no new plugins are planned.

Browse the Cursor plugins archive

Get notified on new tools.

One email per launch. No spam. Unsubscribe anytime.

Frequently asked questions

What is RoninForge? +

A small collection of free, open-source developer tools for people shipping solo. Every tool is MIT licensed, runs locally, and works with the AI coding stack (Claude Code, Cursor). No accounts, no dashboards, no cloud lock-in.

Who is this for? +

Independent developers who ship without a team. People leaning hard on AI coding tools and wanting to keep control of cost, data, and process. Terminal-native, skeptical of vendor lock-in, happy to read source.

Why the name RoninForge? +

A ronin was a samurai without a lord. Shorthand here for working independently by choice, without a company behind you. The tools are built for that audience. The forge is where the work gets hammered out.

Is everything really free and open source? +

Yes. Every repo is MIT licensed. No paid tiers, no subscriptions, no waitlists. If a future tool needs hosted infrastructure that costs us money, it gets labeled clearly and stays opt-in. Nothing shipped today requires that.

Do the tools phone home or collect telemetry? +

No telemetry, no crash reports, no usage analytics in any CLI. The only network calls are ones you configure yourself, like your own ntfy topic for phone alerts. The source is auditable in an afternoon.

How does RoninForge make money then? +

Today, it doesn't. RoninForge exists to build a reputation first. Revenue comes later from a separate paid product built on top of this work, labeled clearly as such. The free tools stay free and MIT.

RoninForge builds free, open-source tools for developers who prefer to own their stack. Everything runs locally. Every repo is MIT licensed. No accounts, no tracking, no vendor lock-in.