Laytime Calculator Pro
AI-powered laytime, demurrage and despatch calculation for ship chartering. Send a Charter Party, Fixture Recap, NOR or Statement of Facts as text or files (PDF, image, Excel) and get a full, auditable laytime statement. Available as a REST API and an MCP server.
README
Laytime Calculator Pro — API & MCP Server
AI-powered laytime, demurrage and despatch calculation for dry-bulk and tanker chartering. Send a Charter Party, Fixture Recap, Notice of Readiness (NOR) or Statement of Facts (SoF) — get back a full, auditable laytime statement with the demurrage / despatch settlement.
Built for software teams in shipping: integrate it into your own systems over a plain REST API, or plug it into an AI assistant via the Model Context Protocol (MCP). Same key, same quota, same engine.
- 🌐 Product: https://navitt.com/web
- 🔑 API access & pricing: https://navitt-mcp.marine-356.workers.dev/authorize
- 📇 MCP Registry:
io.github.shipping-ship-it/laytime-calculator-pro - ✉️ Contact: shipping@agrix.world
What it does
Laytime calculation requires holding every charter-party term in mind at once — laycan, NOR validity, turn time, SHINC/SHEX, weather working days, exceptions, half-rate periods, reversible vs non-reversible time. Miss one clause and the settlement is wrong. This service reads the documents, applies the terms and returns a day-by-day calculation you can audit, with the final demurrage or despatch figure.
- Accepts Charter Party, Fixture Recap, NOR, Statement of Facts, Bills of Lading
- Input as plain text and/or files (PDF, image, Excel)
- Returns a full, line-by-line laytime statement plus the settlement
- Works with any language or platform over REST — no AI required on your side
Quick start (REST)
curl -X POST https://navitt-mcp.marine-356.workers.dev/api/calculate \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"text": "NOR tendered 01 Jun 08:00. Loading 01 Jun 12:00 to 03 Jun 18:00. Laytime allowed 3 days SHINC. Demurrage USD 10000/day. Despatch half demurrage."
}'
Response:
{
"result": "… full day-by-day laytime statement and demurrage/despatch settlement …",
"remaining": 92
}
remaining is the number of calculations left on your plan.
Authentication
Every request must include your API key in the x-api-key header.
Keys are issued after sign-up on the access & pricing page.
x-api-key: YOUR_API_KEY
REST API reference
POST /api/calculate
Base URL: https://navitt-mcp.marine-356.workers.dev
Headers
| Header | Required | Value |
|---|---|---|
x-api-key |
yes | Your API key |
Content-Type |
yes | application/json |
Body (JSON) — provide text, documents, or both:
| Field | Type | Required | Description |
|---|---|---|---|
text |
string | no* | Plain-text input, e.g. a pasted fixture recap or SoF |
documents |
array | no* | Document files, base64-encoded (see below) |
instructions |
string | no | Extra instructions, e.g. "SHEX applies. Demurrage USD 3,500/day" |
* At least one of text or documents must be supplied.
documents[] item
| Field | Type | Description |
|---|---|---|
mimeType |
string | e.g. application/pdf, image/png |
data |
string | base64-encoded file content |
Response 200 OK
| Field | Type | Description |
|---|---|---|
result |
string | The full laytime statement and settlement |
remaining |
integer | Calculations left on your plan |
Errors
| Status | Meaning |
|---|---|
401 |
Missing or invalid x-api-key |
400 |
Body is not valid JSON, or neither text nor documents |
402 |
Calculation quota exhausted (NO_CREDITS) — top up |
502 |
Calculation engine error — retry later |
Each successful request consumes one calculation from your plan.
Example: sending a document (Node.js)
import { readFileSync } from "node:fs";
const data = readFileSync("charter_party.pdf").toString("base64");
const res = await fetch("https://navitt-mcp.marine-356.workers.dev/api/calculate", {
method: "POST",
headers: {
"x-api-key": process.env.LAYTIME_API_KEY,
"Content-Type": "application/json",
},
body: JSON.stringify({
documents: [{ mimeType: "application/pdf", data }],
instructions: "SHINC. Demurrage USD 12,000/day, despatch half demurrage.",
}),
});
const { result, remaining } = await res.json();
console.log(result);
console.log("Calculations left:", remaining);
Example: Python
import base64, requests
with open("statement_of_facts.pdf", "rb") as f:
data = base64.b64encode(f.read()).decode()
r = requests.post(
"https://navitt-mcp.marine-356.workers.dev/api/calculate",
headers={"x-api-key": "YOUR_API_KEY"},
json={"documents": [{"mimeType": "application/pdf", "data": data}]},
)
print(r.json()["result"])
MCP server
The same calculator is available as a Model Context Protocol server, so an AI assistant can call it directly as a tool.
- Endpoint:
https://navitt-mcp.marine-356.workers.dev/mcp(streamable HTTP, OAuth-protected) - Registry name:
io.github.shipping-ship-it/laytime-calculator-pro - Tool:
calculate_laytime
Tool input
| Field | Type | Description |
|---|---|---|
text |
string | Plain-text input (recap, SoF, etc.) |
documents |
array | { mimeType, data } — base64-encoded files |
instructions |
string | Optional extra instructions |
Authorize and obtain access on the access & pricing page.
Pricing
Pay-as-you-go: USD 1,000 for 500 calculations (USD 2 per calculation), by card or by invoice to a corporate account. Corporate rates available for teams. See the access & pricing page or contact shipping@agrix.world.
About
Built by AGRIX / Navitt — modern tools for shipping. BIMCO member 181628.
- https://agrix.world
- https://navitt.com
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。