chat-nest-server 1.0.0

package/README.md

@@ -0,0 +1,95 @@
+# chat-nest-server
+
+> Streaming AI backend server for Chat Nest with built-in cost protection and cancellation propagation.
+
+This package exposes an Express-compatible request handler that:
+- Streams AI responses
+- Enforces rate limits and budgets
+- Supports abort propagation
+- Protects against runaway usage
+
+---
+
+## ✨ Features
+
+- Streaming responses over HTTP
+- End-to-end cancellation support
+- Daily token budget enforcement
+- Rate limiting
+- Message trimming
+- Safe retry semantics
+- OpenAI adapter included
+
+---
+
+## 📦 Installation
+
+```bash
+npm install chat-nest-server
+```
+
+## 🚀 Usage
+Express Integration
+
+```
+import express from "express";
+import cors from "cors";
+import { createChatHandler } from "chat-nest-server";
+
+const app = express();
+
+app.use(cors());
+app.use(express.json());
+
+app.post(
+  "/api/chat",
+  createChatHandler({
+    apiKey: process.env.OPENAI_API_KEY!,
+  })
+);
+
+app.listen(3001, () => {
+  console.log("API running on http://localhost:3001");
+});
+```
+
+---
+
+## 🔐 Environment Variables
+
+| Variable       | Description    |
+| -------------- | -------------- |
+| OPENAI_API_KEY | OpenAI API key |
+
+## 💰 Cost Controls
+
+```
+The server enforces:
+
+Maximum tokens per request
+
+Daily token budget
+
+Request rate limiting
+
+Prompt size trimming
+
+Retry classification
+
+This prevents accidental overspending and abuse.
+```
+
+---
+
+## ⚙ Configuration
+Limits can be customized in:
+
+src/config/
+  aiLimits.ts
+  budget.ts
+
+---
+
+## 📄 License
+
+ISC

package/dist/index.d.ts

@@ -0,0 +1,8 @@
+import { Request, Response } from 'express';
+
+type ChatHandlerConfig = {
+    apiKey: string;
+};
+declare function createChatHandler(config: ChatHandlerConfig): (req: Request, res: Response) => Promise<void>;
+
+export { type ChatHandlerConfig, createChatHandler };

package/dist/index.js

@@ -0,0 +1,142 @@
+// src/createChatHandler.ts
+import OpenAI from "openai";
+
+// src/config/aiLimits.ts
+var AI_MODEL = "gpt-4o-mini";
+var AI_LIMITS = {
+  maxOutputTokens: 300,
+  maxMessages: 6,
+  rateLimitWindowMs: 6e4,
+  maxRequestsPerWindow: 30
+};
+
+// src/config/budget.ts
+var BUDGET = {
+  dailyTokenLimit: 7e4,
+  // ~safe for $5 / 3 months
+  maxTokensPerRequest: 600
+  // input + output guard
+};
+
+// src/utils/tokenEstimator.ts
+function estimateTokens(messages) {
+  const text = messages.map((m) => m.content).join(" ");
+  return Math.ceil(text.length / 4);
+}
+
+// src/utils/tokenBudget.ts
+var currentDay = (/* @__PURE__ */ new Date()).toDateString();
+var tokensUsedToday = 0;
+function canSpendTokens(estimated, limit) {
+  const today = (/* @__PURE__ */ new Date()).toDateString();
+  if (today !== currentDay) {
+    currentDay = today;
+    tokensUsedToday = 0;
+  }
+  return tokensUsedToday + estimated <= limit;
+}
+function recordTokenUsage(used) {
+  tokensUsedToday += used;
+}
+
+// src/createChatHandler.ts
+var requestCount = 0;
+var windowStart = Date.now();
+function isRateLimited() {
+  const now = Date.now();
+  if (now - windowStart > AI_LIMITS.rateLimitWindowMs) {
+    windowStart = now;
+    requestCount = 0;
+  }
+  requestCount++;
+  return requestCount > AI_LIMITS.maxRequestsPerWindow;
+}
+function trimMessages(messages) {
+  return messages.slice(-AI_LIMITS.maxMessages);
+}
+function createChatHandler(config) {
+  if (!config.apiKey) {
+    throw new Error("OPENAI_API_KEY is missing");
+  }
+  const client = new OpenAI({
+    apiKey: config.apiKey
+  });
+  return async function handler(req, res) {
+    const abortController = new AbortController();
+    let streamStarted = false;
+    res.on("close", () => {
+      if (streamStarted && !abortController.signal.aborted) {
+        abortController.abort();
+      }
+    });
+    try {
+      const body = req.body;
+      if (!Array.isArray(body.messages)) {
+        res.status(400).json({ error: "Invalid messages payload" });
+        return;
+      }
+      if (isRateLimited()) {
+        res.status(429).json({
+          error: "Rate limit exceeded. Please slow down."
+        });
+        return;
+      }
+      const trimmedMessages = trimMessages(body.messages);
+      const estimatedInputTokens = estimateTokens(trimmedMessages);
+      const estimatedTotalTokens = estimatedInputTokens + AI_LIMITS.maxOutputTokens;
+      if (estimatedTotalTokens > BUDGET.maxTokensPerRequest || !canSpendTokens(estimatedTotalTokens, BUDGET.dailyTokenLimit)) {
+        res.status(429).json({
+          error: "Daily AI budget exceeded. Try again tomorrow."
+        });
+        return;
+      }
+      res.setHeader("Content-Type", "text/plain");
+      res.setHeader("Transfer-Encoding", "chunked");
+      const stream = await client.chat.completions.create(
+        {
+          model: AI_MODEL,
+          stream: true,
+          temperature: 0.7,
+          max_tokens: AI_LIMITS.maxOutputTokens,
+          messages: trimmedMessages.map((m) => ({
+            role: m.role,
+            content: m.content
+          }))
+        },
+        {
+          signal: abortController.signal
+        }
+      );
+      streamStarted = true;
+      try {
+        for await (const chunk of stream) {
+          if (abortController.signal.aborted) {
+            break;
+          }
+          const token = chunk.choices[0]?.delta?.content;
+          if (token) {
+            res.write(token);
+          }
+        }
+      } catch (error) {
+        if (abortController.signal.aborted) {
+          console.log("Stream aborted by client");
+        } else {
+          throw error;
+        }
+      } finally {
+        recordTokenUsage(estimatedTotalTokens);
+        res.end();
+      }
+    } catch (error) {
+      if (error?.name === "AbortError") {
+        return;
+      }
+      console.error("AI error:", error);
+      res.status(500).json({ error: "AI request failed" });
+    }
+  };
+}
+export {
+  createChatHandler
+};

package/package.json

@@ -0,0 +1,40 @@
+{
+	"name": "chat-nest-server",
+	"version": "1.0.0",
+	"type": "module",
+	"main": "dist/index.js",
+	"license": "ISC",
+	"author": "Shivam Shukla",
+	"types": "dist/index.d.ts",
+	"repository": {
+		"type": "git",
+		"url": "https://github.com/shivams10/chat-nest"
+	},
+	"homepage": "https://github.com/shivams10/chat-nest",
+	"bugs": {
+		"url": "https://github.com/shivams10/chat-nest/issues"
+	},
+	"exports": {
+		".": {
+			"import": "./dist/index.js",
+			"types": "./dist/index.d.ts"
+		}
+	},
+	"files": [
+		"dist"
+	],
+	"scripts": {
+		"build": "tsup src/index.ts --format esm --dts",
+		"dev": "tsup src/index.ts --format esm --watch"
+	},
+	"dependencies": {
+		"chat-nest-core": "*",
+		"express": "^5.2.1",
+		"openai": "^6.16.0"
+	},
+	"devDependencies": {
+		"@types/express": "^5.0.6",
+		"tsup": "^8.5.1",
+		"typescript": "^5.9.3"
+	}
+}