EVE Online Companion MCP Server

EVE Online Companion MCP Server

An MCP server that provides Claude with direct access to EVE Online character data, market prices, killboard intelligence, wiki knowledge, and game mechanics. It enables natural language interaction for checking skills, analyzing fittings, setting in-game destinations, and accessing real-time EVE Online information.

Category
Visit Server

README

EVE Online Companion — MCP Server

Ein MCP-Server (Model Context Protocol) fuer Claude Code, der als persoenlicher EVE Online Companion dient. Gibt Claude direkten Zugriff auf Character-Daten, Marktpreise, Killboard-Intel, Wiki-Wissen und mehr — alles ueber natuerliche Sprache.

Was kann das?

"Kann ich die Ishtar fliegen?" — Prueft deine Skills und zeigt fehlende mit Trainingszeit.

"Was kostet ein Warp Disruptor II in Jita?" — Holt Live-Marktdaten von ESI.

"Wie gefaehrlich ist der Pilot XY?" — Checkt zKillboard-Stats, Danger-Rating, Top-Ships.

"Zeig mir meine Fittings fuer die Vexor" — Listet gespeicherte Fits im EFT-Format.

"Erklaer mir Wormhole-Mechaniken" — Durchsucht die EVE University Wiki.

Alle 29 Tools

Kategorie Tools Beschreibung
Auth eve_auth_login SSO-Login (blockiert bis Callback)
eve_auth_start SSO-Login (non-blocking, oeffne localhost:8834)
eve_auth_status Auth-Status + Character-Info
Character eve_character_info Name, Corp, Alliance, Wallet, SP, Location, Ship
eve_character_skills Skills filtern nach Gruppe oder Name
eve_character_skillqueue Aktuelle Skill-Queue mit Zeiten
eve_character_implants Eingesteckte Implants
Location eve_location_current System, Station, Schiff, Online-Status
eve_route_plan Route berechnen (shortest/secure/insecure)
eve_set_destination Autopilot-Ziel im Spiel setzen
Fittings eve_fitting_list Gespeicherte Fittings (EFT-Format)
eve_fitting_save Fitting aus EFT-String speichern
eve_fitting_analyze Fitting-Analyse (Tank, DPS, Cap — Dogma-basiert)
Market eve_market_price Preis-Check (Jita default, andere Regionen moeglich)
eve_market_appraise Itemliste bewerten (Copy-Paste aus dem Spiel)
eve_market_orders Eigene aktive Kauf-/Verkaufsauftraege
Intel eve_intel_character Pilot-Intel: Corp, Kills, Danger-Rating, Top-Ships
eve_intel_corporation Corp-Intel: Members, Alliance, Killboard-Stats
eve_intel_system System-Intel: Kills, Jumps, NPC-Kills, Ratting
eve_killmail_analyze Killmail aufschluesseln: Fit, Attacker, Damage
Assets eve_assets_search Assets nach Name/Typ durchsuchen
eve_wallet_balance Wallet-Balance
eve_wallet_journal Letzte Wallet-Eintraege
Universe eve_type_info Item/Ship-Details mit Dogma-Attributen
eve_system_info System-Details: Sec, Region, Stationen
eve_search Universelle Suche (Chars, Corps, Systems, Items)
Wiki eve_wiki_search EVE University Wiki durchsuchen
eve_wiki_article Wiki-Artikel lesen (als Markdown)
Fleet eve_fleet_info Fleet-Status (wenn in Fleet)

Setup

1. EVE Developer Application

  1. Gehe zu https://developers.eveonline.com/
  2. "Create New Application"
  3. Application Type: Authentication & API Access
  4. Callback URL: http://localhost:8834/callback
  5. Alle Scopes aus der Liste unten auswaehlen
  6. Client ID notieren (Secret Key optional, nur fuer confidential apps)

<details> <summary><strong>Benoetigte ESI Scopes (20)</strong></summary>

esi-skills.read_skills.v1
esi-skills.read_skillqueue.v1
esi-clones.read_implants.v1
esi-assets.read_assets.v1
esi-wallet.read_character_wallet.v1
esi-fittings.read_fittings.v1
esi-fittings.write_fittings.v1
esi-characters.read_standings.v1
esi-killmails.read_killmails.v1
esi-location.read_location.v1
esi-location.read_ship_type.v1
esi-location.read_online.v1
esi-fleets.read_fleet.v1
esi-markets.read_character_orders.v1
esi-contracts.read_character_contracts.v1
esi-mail.read_mail.v1
esi-ui.open_window.v1
esi-ui.write_waypoint.v1
esi-search.search_structures.v1
esi-universe.read_structures.v1

</details>

2. Installation

git clone <repo>
cd eve
npm install
npm run build

3. Konfiguration

mkdir -p ~/.eve-mcp

Erstelle ~/.eve-mcp/config.json:

{
  "clientId": "deine-client-id-von-ccp",
  "callbackUrl": "http://localhost:8834/callback",
  "userAgent": "eve-mcp-companion/1.0 (dein-character-name)"
}

Optional mit Secret Key (nur bei confidential apps):

{
  "clientId": "deine-client-id",
  "secretKey": "dein-secret-key",
  "callbackUrl": "http://localhost:8834/callback",
  "userAgent": "eve-mcp-companion/1.0 (dein-character-name)"
}

Alternativ via Environment-Variablen:

export EVE_MCP_CLIENT_ID="deine-client-id"
export EVE_MCP_CALLBACK_URL="http://localhost:8834/callback"
export EVE_MCP_USER_AGENT="eve-mcp-companion/1.0"

4. Claude Code Integration

Fuege den MCP-Server in deine Claude Code Settings ein.

Option A — settings.json (empfohlen):

Datei: ~/.claude/settings.json

{
  "mcpServers": {
    "eve-online": {
      "command": "node",
      "args": ["/absoluter/pfad/zu/eve/dist/index.js"],
      "env": {
        "EVE_MCP_TOKEN_PASSPHRASE": "ein-sicheres-passwort"
      }
    }
  }
}

Option B — Projekt-scope (.mcp.json im Projektverzeichnis):

{
  "mcpServers": {
    "eve-online": {
      "command": "node",
      "args": ["./dist/index.js"],
      "env": {
        "EVE_MCP_TOKEN_PASSPHRASE": "ein-sicheres-passwort"
      }
    }
  }
}

5. Erster Login

Starte Claude Code und sage:

"Verbinde mich mit EVE Online"

Claude ruft eve_auth_start auf und startet einen lokalen Auth-Server. Oeffne http://localhost:8834 im Browser — dort erscheint eine Login-Seite im EVE-Stil. Klicke "Authenticate via EVE SSO", logge dich bei CCP ein, und du wirst zurueckgeleitet zu einer Erfolgsseite mit deinem Character-Portrait. Fenster schliessen, fertig.


Architektur

eve/
├── src/
│   ├── index.ts                 # Server-Entry, registriert alle Tools
│   ├── config.ts                # Laedt ~/.eve-mcp/config.json
│   ├── auth/
│   │   ├── scopes.ts            # ESI Scope-Definitionen
│   │   ├── sso.ts               # OAuth2 PKCE Flow, Callback-Server
│   │   ├── tokens.ts            # Token-Persistence (AES-256-GCM)
│   │   └── pages.ts             # HTML-Seiten fuer Auth-Flow
│   ├── clients/
│   │   ├── esi.ts               # ESI API Client (Auth, Rate-Limiting, Cache)
│   │   ├── zkillboard.ts        # zKillboard Client (10 req/s Throttle)
│   │   ├── evetycoon.ts         # EVE Tycoon Markt-Client
│   │   └── wiki.ts              # EVE University Wiki (MediaWiki API)
│   ├── tools/
│   │   ├── auth.ts              # Login, Status
│   │   ├── character.ts         # Character Info, Skills, Queue, Implants
│   │   ├── location.ts          # Location, Route, Autopilot
│   │   ├── fitting.ts           # Fittings, EFT, Analyse
│   │   ├── universe.ts          # Type Info, System Info, Search
│   │   ├── market.ts            # Preise, Appraisal, Orders
│   │   ├── killboard.ts         # Character/Corp/System Intel, Killmails
│   │   ├── assets.ts            # Assets, Wallet
│   │   ├── wiki.ts              # Wiki Search, Article
│   │   └── fleet.ts             # Fleet Info
│   └── utils/
│       ├── cache.ts             # In-Memory Cache mit TTL
│       ├── errors.ts            # Error-Klassen + formatToolError()
│       ├── formatting.ts        # ISK, Zeit, EVE-Time Formatierung
│       ├── eft.ts               # EFT-Format Parser + Generator
│       └── sde.ts               # Type/System Name-Resolution (ESI-backed)
├── package.json
├── tsconfig.json
└── README.md

API-Clients

Client Base URL Auth Rate Limit Cache
ESI esi.evetech.net/latest OAuth2 Bearer Error-Limit Header Expires-Header + custom TTL
zKillboard zkillboard.com/api keine 10 req/s 10 Minuten
EVE Tycoon evetycoon.com/api/v1 keine Expires-Header 5 Minuten
Wiki wiki.eveuniversity.org/api.php keine keine 1 Stunde

Caching

Alle API-Antworten werden in-memory gecacht:

Datentyp TTL
Location / Ship / Online 30 Sekunden
Wallet Balance 2 Minuten
Skill Queue 5 Minuten
Market Prices / Orders 5 Minuten
zKillboard Stats 10 Minuten
Assets 30 Minuten
Character Skills 1 Stunde
Wiki Articles 1 Stunde
Type Info / System Info 24 Stunden
Name-zu-ID Resolution 24 Stunden

ESI Rate-Limiting

Der ESI-Client trackt X-ESI-Error-Limit-Remain und X-ESI-Error-Limit-Reset Header. Wenn das Error-Limit unter 20 faellt, werden Requests bis zum Reset blockiert. Fehler 420 (Error Limit) werden als RateLimitError geworfen.

Token-Sicherheit

  • Tokens werden in ~/.eve-mcp/tokens.json gespeichert
  • Wenn EVE_MCP_TOKEN_PASSPHRASE gesetzt ist: AES-256-GCM Verschluesselung mit scrypt-Key-Derivation
  • Ohne Passphrase: Klartext-Speicherung (mit Warnung auf stderr)
  • Access-Token wird automatisch 2 Minuten vor Ablauf refreshed
  • Refresh-Token wird bei jedem Refresh erneuert (PKCE volatile token pattern)

Entwicklung

# Dev-Mode (tsx, kein Build noetig)
npm run dev

# Build
npm run build

# Ausfuehren
npm start

Neues Tool hinzufuegen

  1. Handler in die passende Datei unter src/tools/ einfuegen
  2. server.tool(name, description, zodSchema, handler) Pattern verwenden
  3. Fehler immer mit formatToolError(err) zurueckgeben
  4. ESI-Calls ueber esi.get() / esi.publicGet() mit passendem Cache-TTL
  5. Type-IDs mit resolveTypeName() / resolveTypeId() aufloesen

Troubleshooting

"Config not found"~/.eve-mcp/config.json anlegen oder EVE_MCP_CLIENT_ID setzen.

"Authentication required"eve_auth_login oder eve_auth_start aufrufen.

"Token refresh failed" — Refresh-Token abgelaufen. Erneut einloggen.

"ESI error limit reached" — Zu viele fehlerhafte Requests. Wartet automatisch.

Port 8834 belegt — Anderen Port in config.json setzen und Callback-URL bei CCP anpassen.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured