octopus-mcp

octopus-mcp

Lets Claude analyse your Octopus Energy account: usage, costs, tariff comparisons, Saving Sessions, Octoplus rewards.

Category
访问服务器

README

octopus-mcp

PyPI CI

Unofficial. Not affiliated with Octopus Energy. Uses the public REST API and the community-known Kraken GraphQL endpoint; the latter is unofficial and may break without notice.

A Model Context Protocol server that lets Claude analyse your Octopus Energy account: usage, costs, tariff comparisons, Saving Sessions, Octoplus rewards.

Works with Claude Code, Claude Desktop, and any MCP-compatible client.

Prerequisites

  • Python 3.11 or newer
  • An Octopus Energy account in the UK
  • A smart meter sending half-hourly readings (so consumption data is available via the API)

Install

The server is published on PyPI as octopus-mcp. You can install it with uv (recommended) or plain pip.

With uv (recommended)

If you don't have uv yet:

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Then install the MCP server and save credentials:

uv tool install octopus-mcp
octopus-mcp configure   # interactive; writes to your OS keychain

With pip

pip install octopus-mcp
octopus-mcp configure

Wire it into your MCP client

Claude Code

In Claude Code, register this repo as a plugin marketplace and install the octopus plugin from it:

/plugin marketplace add DanielChicot/octopus-mcp
/plugin install octopus@octopus-mcp

The plugin auto-registers the MCP server (no manual config-file edits needed) and adds four slash commands (/octopus:bill, /octopus:compare, /octopus:peaks, /octopus:saving-sessions) plus an analysis skill that Claude triggers automatically on energy-related questions.

Claude Desktop, Cursor, and other MCP clients

Add this entry to your client's MCP config:

{
  "mcpServers": {
    "octopus": {
      "command": "uvx",
      "args": ["octopus-mcp"]
    }
  }
}

(If you installed with pip instead of uv, replace "command": "uvx", "args": ["octopus-mcp"] with "command": "octopus-mcp", "args": ["serve"].)

The config file lives at:

Platform Path
macOS ~/Library/Application Support/Claude/claude_desktop_config.json
Linux ~/.config/Claude/claude_desktop_config.json
Windows %APPDATA%\Claude\claude_desktop_config.json

Restart your MCP client after editing the config.

Credentials

You need:

Optional (only some Kraken queries):

  • OCTOPUS_EMAIL / OCTOPUS_PASSWORD

Resolution order: shell env > .env > OS keyring > error. The recommended path is octopus-mcp configure which writes to the OS keyring so secrets never sit in plaintext config.

Tools

Tool What it does
bill_summary(period) Total kWh and £ per fuel for a period
usage_breakdown(period, group_by) Aggregated kWh by hour/day/week/month
peak_hours(period, top_n) Top-N highest-usage half-hours
compare_tariff(target_product_code, period, fuel) Replay your usage against another Octopus tariff
current_tariff() What you're on now: unit rate, standing charge
saving_session_history() Octoplus saving sessions joined and rewards earned
get_account(), list_products(), get_product(), get_consumption_raw(), kraken_query() Thin getters / escape hatches

Slash commands (Claude Code plugin)

  • /octopus:bill [period] — bill summary as a markdown table
  • /octopus:compare <product-code> [period] — tariff comparison with caveats
  • /octopus:peaks [period] [top_n] — highest-usage half-hours
  • /octopus:saving-sessions — Octoplus history

How it works

  • A SQLite cache at ~/Library/Caches/octopus-mcp/ (or your platform's equivalent) holds your historical consumption and tariff data, refreshed incrementally.
  • All cost figures are in integer pence inc-VAT (no float drift), with a derived pounds string for display.
  • Tariff comparison is a pure tariff swap model — caveats list what it does and doesn't model. See the design doc.

Known limitations (v0.1)

  • Gas SMETS2 m³ vs kWh: if your gas meter reports in m³, values won't be normalised — expect implausibly small numbers and multiply by ~11.18 to convert. v0.2 will detect and apply calorific conversion automatically.
  • Region-aware tariff lookup: compare_tariff currently picks the first region's tariff variant from a target product. v0.2 will use your postcode-derived region.
  • TTLs are hardcoded: no config.toml support yet.
  • No background sync: consumption is fetched lazily on demand.

Privacy

The MCP runs on your machine. No data leaves your computer except direct calls to api.octopus.energy. Credentials live in your OS keychain. Logs at ~/Library/Logs/octopus-mcp/server.log redact secrets.

Contributing

See CONTRIBUTING.md. Run pre-commit run --all-files before pushing.

License

MIT — see LICENSE.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选