feat(model-service): add OpenAI-compatible wrapper (+ pm2 + env example) and update ignores

.gitignore

@@ -158,3 +158,5 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+model-service/.env
+

model-service/.env.example

@@ -0,0 +1,5 @@
+OPENAI_BASE_URL=https://api.openai.com/v1
+OPENAI_API_KEY=sk-REPLACE_ME
+MODEL_NAME=gpt-4o-mini
+PORT=7071
+

model-service/.gitignore

@@ -0,0 +1,5 @@
+package-lock.json
+node_modules/
+.env
+.DS_Store
+

model-service/Procfile

@@ -0,0 +1,2 @@
+web: node server.js
+

model-service/ecosystem.config.js

@@ -0,0 +1,16 @@
+module.exports = {
+  apps: [
+    {
+      name: 'reentry-model-service',
+      script: 'server.js',
+      instances: 1,
+      exec_mode: 'fork',
+      env: {
+        // server.js loads dotenv; these are optional overrides
+        PORT: process.env.PORT || 7071,
+      },
+      watch: false,
+    },
+  ],
+};
+

model-service/package.json

@@ -0,0 +1,20 @@
+{
+  "name": "reentry-model-service",
+  "version": "1.0.0",
+  "private": true,
+  "main": "server.js",
+  "license": "UNLICENSED",
+  "scripts": {
+    "dev": "node server.js",
+    "start": "node server.js",
+    "pm2": "pm2 start ecosystem.config.js"
+  },
+  "dependencies": {
+    "dotenv": "^16.4.5",
+    "express": "^4.19.2",
+    "node-fetch": "^2.7.0"
+  },
+  "engines": {
+    "node": ">=18"
+  }
+}

model-service/server.js

@@ -0,0 +1,64 @@
+const express = require('express');
+const fetch = require('node-fetch');
+const dotenv = require('dotenv');
+
+dotenv.config();
+
+const app = express();
+app.use(express.json());
+
+const PORT = process.env.PORT || 7071;
+const OPENAI_BASE_URL = (process.env.OPENAI_BASE_URL || 'https://api.openai.com/v1').replace(/\/+$/, '');
+const OPENAI_API_KEY = process.env.OPENAI_API_KEY || '';
+const MODEL_NAME = process.env.MODEL_NAME || 'gpt-4o-mini';
+
+// Health endpoint
+app.get('/health', (_req, res) => {
+  res.json({ ok: true, model: MODEL_NAME, base: OPENAI_BASE_URL });
+});
+
+// Generate endpoint - forwards to provider-compatible chat/completions
+app.post('/generate', async (req, res) => {
+  try {
+    if (!OPENAI_API_KEY) {
+      return res.status(400).json({ error: 'Missing OPENAI_API_KEY in environment' });
+    }
+
+    const body = req.body || {};
+    const messages = body.messages;
+    const model = body.model || MODEL_NAME;
+    const extra = body.extra || {}; // allow optional pass-through params
+
+    if (!Array.isArray(messages) || messages.length === 0) {
+      return res.status(400).json({ error: 'Body must include messages: [{ role, content }]' });
+    }
+
+    const url = `${OPENAI_BASE_URL}/chat/completions`;
+
+    const providerResp = await fetch(url, {
+      method: 'POST',
+      headers: {
+        'Authorization': `Bearer ${OPENAI_API_KEY}`,
+        'Content-Type': 'application/json'
+      },
+      body: JSON.stringify({ model, messages, ...extra })
+    });
+
+    const data = await providerResp.json();
+
+    if (!providerResp.ok) {
+      // Pass through provider error
+      return res.status(providerResp.status).json({ error: data.error || data });
+    }
+
+    const text = data?.choices?.[0]?.message?.content || '';
+    return res.json({ text });
+  } catch (err) {
+    return res.status(500).json({ error: 'Upstream error', detail: String(err && err.message || err) });
+  }
+});
+
+app.listen(PORT, () => {
+  console.log(`model-service listening on http://localhost:${PORT}`);
+});
+