free-coding-models 0.1.82 → 0.1.84

package/src/opencode.js

@@ -0,0 +1,952 @@
+/**
+ * @file opencode.js
+ * @description OpenCode integration and multi-account proxy lifecycle.
+ *
+ * @details
+ *   This module owns all OpenCode-related behavior:
+ *   - Configure opencode.json with selected models/providers
+ *   - Launch OpenCode CLI or Desktop
+ *   - Manage ZAI proxy bridge for non-standard API paths
+ *   - Start/stop the multi-account proxy server (fcm-proxy)
+ *   - Auto-start proxy when OpenCode config is already synced
+ *
+ *   🎯 Key features:
+ *   - Provider-aware config setup for OpenCode (NIM, Groq, Cerebras, etc.)
+ *   - ZAI proxy bridge to rewrite /v1/* → /api/coding/paas/v4/*
+ *   - Auto-pick tmux port for OpenCode sub-agents
+ *   - Multi-account proxy with rotation + auto-sync to opencode.json
+ *
+ *   → Functions:
+ *   - `setOpenCodeModelData` — Provide merged model lists for proxy topology
+ *   - `startOpenCode` — Launch OpenCode CLI with selected model
+ *   - `startOpenCodeDesktop` — Set model and open Desktop app
+ *   - `startProxyAndLaunch` — Start fcm-proxy then launch OpenCode
+ *   - `autoStartProxyIfSynced` — Auto-start proxy if opencode.json has fcm-proxy
+ *   - `ensureProxyRunning` — Ensure proxy is running (start or reuse)
+ *
+ *   @see src/opencode-sync.js — syncToOpenCode/load/save utilities
+ *   @see src/proxy-server.js  — ProxyServer implementation
+ */
+
+import chalk from 'chalk'
+import { createServer } from 'net'
+import { createServer as createHttpServer } from 'http'
+import { request as httpsRequest } from 'https'
+import { randomUUID } from 'crypto'
+import { homedir } from 'os'
+import { join } from 'path'
+import { copyFileSync, existsSync } from 'fs'
+import { sources } from '../sources.js'
+import { resolveCloudflareUrl } from './ping.js'
+import { ProxyServer } from './proxy-server.js'
+import { loadOpenCodeConfig, saveOpenCodeConfig, syncToOpenCode } from './opencode-sync.js'
+import { getApiKey, resolveApiKeys } from './config.js'
+import { ENV_VAR_NAMES, OPENCODE_MODEL_MAP, isWindows, isMac, isLinux } from './provider-metadata.js'
+import { setActiveProxy } from './render-table.js'
+
+// 📖 OpenCode config location: ~/.config/opencode/opencode.json on ALL platforms.
+// 📖 OpenCode uses xdg-basedir which resolves to %USERPROFILE%\.config on Windows.
+const OPENCODE_CONFIG = join(homedir(), '.config', 'opencode', 'opencode.json')
+const OPENCODE_PORT_RANGE_START = 4096
+const OPENCODE_PORT_RANGE_END = 5096
+
+// 📖 Module-level proxy state — shared between startProxyAndLaunch and cleanup.
+let activeProxy = null
+let proxyCleanedUp = false
+let exitHandlersRegistered = false
+
+// 📖 Merged model references for proxy topology.
+let mergedModelsRef = []
+let mergedModelByLabelRef = new Map()
+
+// 📖 setOpenCodeModelData: Provide mergedModels + mergedModelByLabel to this module.
+export function setOpenCodeModelData(mergedModels, mergedModelByLabel) {
+  mergedModelsRef = Array.isArray(mergedModels) ? mergedModels : []
+  mergedModelByLabelRef = mergedModelByLabel instanceof Map ? mergedModelByLabel : new Map()
+}
+
+// 📖 isTcpPortAvailable: checks if a local TCP port is free for OpenCode.
+// 📖 Used to avoid tmux sub-agent port conflicts when multiple projects run in parallel.
+function isTcpPortAvailable(port) {
+  return new Promise((resolve) => {
+    const server = createServer()
+    server.once('error', () => resolve(false))
+    server.once('listening', () => {
+      server.close(() => resolve(true))
+    })
+    server.listen(port)
+  })
+}
+
+// 📖 resolveOpenCodeTmuxPort: selects a safe port for OpenCode when inside tmux.
+// 📖 Priority:
+// 📖 1) OPENCODE_PORT from env (if valid and available)
+// 📖 2) First available port in 4096-5095
+async function resolveOpenCodeTmuxPort() {
+  const envPortRaw = process.env.OPENCODE_PORT
+  const envPort = Number.parseInt(envPortRaw || '', 10)
+
+  if (Number.isInteger(envPort) && envPort > 0 && envPort <= 65535) {
+    if (await isTcpPortAvailable(envPort)) {
+      return { port: envPort, source: 'env' }
+    }
+    console.log(chalk.yellow(`  ⚠ OPENCODE_PORT=${envPort} is already in use; selecting another port for this run.`))
+  }
+
+  for (let port = OPENCODE_PORT_RANGE_START; port < OPENCODE_PORT_RANGE_END; port++) {
+    if (await isTcpPortAvailable(port)) {
+      return { port, source: 'auto' }
+    }
+  }
+
+  return null
+}
+
+function getOpenCodeConfigPath() {
+  return OPENCODE_CONFIG
+}
+
+// 📖 Map source model IDs to OpenCode built-in IDs when they differ.
+function getOpenCodeModelId(providerKey, modelId) {
+  if (providerKey === 'nvidia') return modelId.replace(/^nvidia\//, '')
+  if (providerKey === 'zai') return modelId.replace(/^zai\//, '')
+  return OPENCODE_MODEL_MAP[providerKey]?.[modelId] || modelId
+}
+
+// ─── ZAI proxy bridge ─────────────────────────────────────────────────────────
+
+// 📖 createZaiProxy: Localhost reverse proxy bridging ZAI's non-standard API paths
+// 📖 to OpenCode's expected /v1/* OpenAI-compatible format.
+// 📖 Returns { server, port } — caller must server.close() when done.
+async function createZaiProxy(apiKey) {
+  const server = createHttpServer((req, res) => {
+    let targetPath = req.url
+    // 📖 Rewrite /v1/* → /api/coding/paas/v4/*
+    if (targetPath.startsWith('/v1/')) {
+      targetPath = '/api/coding/paas/v4/' + targetPath.slice(4)
+    } else if (targetPath.startsWith('/v1')) {
+      targetPath = '/api/coding/paas/v4' + targetPath.slice(3)
+    } else {
+      // 📖 Non /v1 paths (e.g. /api/v0/ health checks) — reject
+      res.writeHead(404)
+      res.end()
+      return
+    }
+    const headers = { ...req.headers, host: 'api.z.ai' }
+    if (apiKey) headers.authorization = `Bearer ${apiKey}`
+    // 📖 Remove transfer-encoding to avoid chunked encoding issues with https.request
+    delete headers['transfer-encoding']
+    const proxyReq = httpsRequest({
+      hostname: 'api.z.ai',
+      port: 443,
+      path: targetPath,
+      method: req.method,
+      headers,
+    }, (proxyRes) => {
+      res.writeHead(proxyRes.statusCode, proxyRes.headers)
+      proxyRes.pipe(res)
+    })
+    proxyReq.on('error', () => { res.writeHead(502); res.end() })
+    req.pipe(proxyReq)
+  })
+  await new Promise(r => server.listen(0, '127.0.0.1', r))
+  return { server, port: server.address().port }
+}
+
+// ─── Shared OpenCode spawn helper ─────────────────────────────────────────────
+
+// 📖 spawnOpenCode: Resolve API keys + spawn opencode CLI with correct env.
+async function spawnOpenCode(args, providerKey, fcmConfig, existingZaiProxy = null) {
+  const envVarName = ENV_VAR_NAMES[providerKey]
+  const resolvedKey = getApiKey(fcmConfig, providerKey)
+  const childEnv = { ...process.env }
+  // 📖 Suppress MaxListenersExceededWarning from @modelcontextprotocol/sdk
+  childEnv.NODE_NO_WARNINGS = '1'
+  const finalArgs = [...args]
+  const hasExplicitPortArg = finalArgs.includes('--port')
+  if (envVarName && resolvedKey) childEnv[envVarName] = resolvedKey
+
+  // 📖 ZAI proxy: OpenCode's Go binary doesn't know about ZAI as a provider.
+  // 📖 Start proxy if needed, or reuse existing proxy if passed in.
+  let zaiProxy = existingZaiProxy
+  if (providerKey === 'zai' && resolvedKey && !zaiProxy) {
+    const { server, port } = await createZaiProxy(resolvedKey)
+    zaiProxy = server
+    console.log(chalk.dim(`  🔀 ZAI proxy listening on port ${port} (rewrites /v1/* → ZAI API)`))
+  }
+
+  // 📖 In tmux, OpenCode sub-agents need a listening port to open extra panes.
+  if (process.env.TMUX && !hasExplicitPortArg) {
+    const tmuxPort = await resolveOpenCodeTmuxPort()
+    if (tmuxPort) {
+      const portValue = String(tmuxPort.port)
+      childEnv.OPENCODE_PORT = portValue
+      finalArgs.push('--port', portValue)
+      if (tmuxPort.source === 'env') {
+        console.log(chalk.dim(`  📺 tmux detected — using OPENCODE_PORT=${portValue}.`))
+      } else {
+        console.log(chalk.dim(`  📺 tmux detected — using OpenCode port ${portValue} for sub-agent panes.`))
+      }
+    } else {
+      console.log(chalk.yellow(`  ⚠ tmux detected but no free OpenCode port found in ${OPENCODE_PORT_RANGE_START}-${OPENCODE_PORT_RANGE_END - 1}; launching without --port.`))
+    }
+  }
+
+  const { spawn } = await import('child_process')
+  const child = spawn('opencode', finalArgs, {
+    stdio: 'inherit',
+    shell: true,
+    detached: false,
+    env: childEnv
+  })
+
+  return new Promise((resolve, reject) => {
+    child.on('exit', (code) => {
+      if (zaiProxy) zaiProxy.close()
+      // 📖 ZAI cleanup: remove the ephemeral proxy provider from opencode.json
+      if (providerKey === 'zai') {
+        try {
+          const cfg = loadOpenCodeConfig()
+          if (cfg.provider?.zai) delete cfg.provider.zai
+          if (typeof cfg.model === 'string' && cfg.model.startsWith('zai/')) delete cfg.model
+          saveOpenCodeConfig(cfg)
+        } catch { /* best-effort cleanup */ }
+      }
+      resolve(code)
+    })
+    child.on('error', (err) => {
+      if (zaiProxy) zaiProxy.close()
+      if (err.code === 'ENOENT') {
+        console.error(chalk.red('\n  X Could not find "opencode" -- is it installed and in your PATH?'))
+        console.error(chalk.dim('    Install: npm i -g opencode   or see https://opencode.ai'))
+        resolve(1)
+      } else {
+        reject(err)
+      }
+    })
+  })
+}
+
+// ─── Start OpenCode CLI ───────────────────────────────────────────────────────
+
+export async function startOpenCode(model, fcmConfig) {
+  const providerKey = model.providerKey ?? 'nvidia'
+  const ocModelId = getOpenCodeModelId(providerKey, model.modelId)
+  const modelRef = `${providerKey}/${ocModelId}`
+
+  if (providerKey === 'nvidia') {
+    const config = loadOpenCodeConfig()
+    const backupPath = `${getOpenCodeConfigPath()}.backup-${Date.now()}`
+
+    if (existsSync(getOpenCodeConfigPath())) {
+      copyFileSync(getOpenCodeConfigPath(), backupPath)
+      console.log(chalk.dim(`  Backup: ${backupPath}`))
+    }
+
+    if (!config.provider) config.provider = {}
+    if (!config.provider.nvidia) {
+      config.provider.nvidia = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'NVIDIA NIM',
+        options: {
+          baseURL: 'https://integrate.api.nvidia.com/v1',
+          apiKey: '{env:NVIDIA_API_KEY}'
+        },
+        models: {}
+      }
+      console.log(chalk.green('  + Auto-configured NVIDIA NIM provider in OpenCode'))
+    }
+
+    console.log(chalk.green(`  Setting ${chalk.bold(model.label)} as default...`))
+    console.log(chalk.dim(`  Model: ${modelRef}`))
+    console.log()
+
+    config.model = modelRef
+    if (!config.provider.nvidia.models) config.provider.nvidia.models = {}
+    config.provider.nvidia.models[ocModelId] = { name: model.label }
+
+    saveOpenCodeConfig(config)
+
+    const savedConfig = loadOpenCodeConfig()
+    console.log(chalk.dim(`  Config saved to: ${getOpenCodeConfigPath()}`))
+    console.log(chalk.dim(`  Default model in config: ${savedConfig.model || 'NOT SET'}`))
+    console.log()
+
+    if (savedConfig.model === config.model) {
+      console.log(chalk.green(`  Default model set to: ${modelRef}`))
+    } else {
+      console.log(chalk.yellow(`  Config might not have been saved correctly`))
+    }
+    console.log()
+    console.log(chalk.dim('  Starting OpenCode...'))
+    console.log()
+
+    await spawnOpenCode(['--model', modelRef], providerKey, fcmConfig)
+    return
+  }
+
+  if (providerKey === 'replicate') {
+    console.log(chalk.yellow('  Replicate models are monitor-only for now in OpenCode mode.'))
+    console.log(chalk.dim('    Reason: Replicate uses /v1/predictions instead of OpenAI chat-completions.'))
+    console.log(chalk.dim('    You can still benchmark this model in the TUI and use other providers for OpenCode launch.'))
+    console.log()
+    return
+  }
+
+  if (providerKey === 'zai') {
+    const resolvedKey = getApiKey(fcmConfig, providerKey)
+    if (!resolvedKey) {
+      console.log(chalk.yellow('  ZAI API key not found. Set ZAI_API_KEY environment variable.'))
+      console.log()
+      return
+    }
+
+    const { server: zaiProxyServer, port: zaiProxyPort } = await createZaiProxy(resolvedKey)
+    console.log(chalk.dim(`  ZAI proxy listening on port ${zaiProxyPort} (rewrites /v1/* -> ZAI API)`))
+
+    console.log(chalk.green(`  Setting ${chalk.bold(model.label)} as default...`))
+    console.log(chalk.dim(`  Model: ${modelRef}`))
+    console.log()
+
+    const config = loadOpenCodeConfig()
+    const backupPath = `${getOpenCodeConfigPath()}.backup-${Date.now()}`
+
+    if (existsSync(getOpenCodeConfigPath())) {
+      copyFileSync(getOpenCodeConfigPath(), backupPath)
+      console.log(chalk.dim(`  Backup: ${backupPath}`))
+    }
+
+    if (!config.provider) config.provider = {}
+    config.provider.zai = {
+      npm: '@ai-sdk/openai-compatible',
+      name: 'ZAI',
+      options: {
+        baseURL: `http://127.0.0.1:${zaiProxyPort}/v1`,
+        apiKey: 'zai-proxy',
+      },
+      models: {}
+    }
+    config.provider.zai.models[ocModelId] = { name: model.label }
+    config.model = modelRef
+
+    saveOpenCodeConfig(config)
+
+    const savedConfig = loadOpenCodeConfig()
+    console.log(chalk.dim(`  Config saved to: ${getOpenCodeConfigPath()}`))
+    console.log(chalk.dim(`  Default model in config: ${savedConfig.model || 'NOT SET'}`))
+    console.log()
+
+    if (savedConfig.model === config.model) {
+      console.log(chalk.green(`  Default model set to: ${modelRef}`))
+    } else {
+      console.log(chalk.yellow(`  Config might not have been saved correctly`))
+    }
+    console.log()
+    console.log(chalk.dim('  Starting OpenCode...'))
+    console.log()
+
+    await spawnOpenCode(['--model', modelRef], providerKey, fcmConfig, zaiProxyServer)
+    return
+  }
+
+  console.log(chalk.green(`  Setting ${chalk.bold(model.label)} as default...`))
+  console.log(chalk.dim(`  Model: ${modelRef}`))
+  console.log()
+
+  const config = loadOpenCodeConfig()
+  const backupPath = `${getOpenCodeConfigPath()}.backup-${Date.now()}`
+
+  if (existsSync(getOpenCodeConfigPath())) {
+    copyFileSync(getOpenCodeConfigPath(), backupPath)
+    console.log(chalk.dim(`  Backup: ${backupPath}`))
+  }
+
+  if (!config.provider) config.provider = {}
+  if (!config.provider[providerKey]) {
+    if (providerKey === 'groq') {
+      config.provider.groq = { options: { apiKey: '{env:GROQ_API_KEY}' }, models: {} }
+    } else if (providerKey === 'cerebras') {
+      config.provider.cerebras = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Cerebras',
+        options: { baseURL: 'https://api.cerebras.ai/v1', apiKey: '{env:CEREBRAS_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'sambanova') {
+      config.provider.sambanova = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'SambaNova',
+        options: { baseURL: 'https://api.sambanova.ai/v1', apiKey: '{env:SAMBANOVA_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'openrouter') {
+      config.provider.openrouter = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'OpenRouter',
+        options: { baseURL: 'https://openrouter.ai/api/v1', apiKey: '{env:OPENROUTER_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'huggingface') {
+      config.provider.huggingface = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Hugging Face Inference',
+        options: { baseURL: 'https://router.huggingface.co/v1', apiKey: '{env:HUGGINGFACE_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'deepinfra') {
+      config.provider.deepinfra = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'DeepInfra',
+        options: { baseURL: 'https://api.deepinfra.com/v1/openai', apiKey: '{env:DEEPINFRA_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'fireworks') {
+      config.provider.fireworks = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Fireworks AI',
+        options: { baseURL: 'https://api.fireworks.ai/inference/v1', apiKey: '{env:FIREWORKS_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'codestral') {
+      config.provider.codestral = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Mistral Codestral',
+        options: { baseURL: 'https://codestral.mistral.ai/v1', apiKey: '{env:CODESTRAL_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'hyperbolic') {
+      config.provider.hyperbolic = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Hyperbolic',
+        options: { baseURL: 'https://api.hyperbolic.xyz/v1', apiKey: '{env:HYPERBOLIC_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'scaleway') {
+      config.provider.scaleway = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Scaleway',
+        options: { baseURL: 'https://api.scaleway.ai/v1', apiKey: '{env:SCALEWAY_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'googleai') {
+      config.provider.googleai = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Google AI Studio',
+        options: { baseURL: 'https://generativelanguage.googleapis.com/v1beta/openai', apiKey: '{env:GOOGLE_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'siliconflow') {
+      config.provider.siliconflow = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'SiliconFlow',
+        options: { baseURL: 'https://api.siliconflow.com/v1', apiKey: '{env:SILICONFLOW_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'together') {
+      config.provider.together = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Together AI',
+        options: { baseURL: 'https://api.together.xyz/v1', apiKey: '{env:TOGETHER_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'cloudflare') {
+      const cloudflareAccountId = (process.env.CLOUDFLARE_ACCOUNT_ID || '').trim()
+      if (!cloudflareAccountId) {
+        console.log(chalk.yellow('  Cloudflare Workers AI requires CLOUDFLARE_ACCOUNT_ID for OpenCode integration.'))
+        console.log(chalk.dim('    Export CLOUDFLARE_ACCOUNT_ID and retry this selection.'))
+        console.log()
+        return
+      }
+      config.provider.cloudflare = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Cloudflare Workers AI',
+        options: { baseURL: `https://api.cloudflare.com/client/v4/accounts/${cloudflareAccountId}/ai/v1`, apiKey: '{env:CLOUDFLARE_API_TOKEN}' },
+        models: {}
+      }
+    } else if (providerKey === 'perplexity') {
+      config.provider.perplexity = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Perplexity API',
+        options: { baseURL: 'https://api.perplexity.ai', apiKey: '{env:PERPLEXITY_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'iflow') {
+      config.provider.iflow = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'iFlow',
+        options: { baseURL: 'https://apis.iflow.cn/v1', apiKey: '{env:IFLOW_API_KEY}' },
+        models: {}
+      }
+    }
+  }
+
+  const isBuiltinMapped = OPENCODE_MODEL_MAP[providerKey]?.[model.modelId]
+  if (!isBuiltinMapped) {
+    if (!config.provider[providerKey].models) config.provider[providerKey].models = {}
+    config.provider[providerKey].models[ocModelId] = { name: model.label }
+  }
+
+  config.model = modelRef
+  saveOpenCodeConfig(config)
+
+  const savedConfig = loadOpenCodeConfig()
+  console.log(chalk.dim(`  Config saved to: ${getOpenCodeConfigPath()}`))
+  console.log(chalk.dim(`  Default model in config: ${savedConfig.model || 'NOT SET'}`))
+  console.log()
+
+  if (savedConfig.model === config.model) {
+    console.log(chalk.green(`  Default model set to: ${modelRef}`))
+  } else {
+    console.log(chalk.yellow(`  Config might not have been saved correctly`))
+  }
+  console.log()
+  console.log(chalk.dim('  Starting OpenCode...'))
+  console.log()
+
+  await spawnOpenCode(['--model', modelRef], providerKey, fcmConfig)
+}
+
+// ─── Proxy lifecycle ─────────────────────────────────────────────────────────
+
+async function cleanupProxy() {
+  if (proxyCleanedUp || !activeProxy) return
+  proxyCleanedUp = true
+  const proxy = activeProxy
+  activeProxy = null
+  setActiveProxy(activeProxy)
+  try {
+    await proxy.stop()
+  } catch { /* best-effort */ }
+}
+
+function registerExitHandlers() {
+  if (exitHandlersRegistered) return
+  exitHandlersRegistered = true
+  const cleanup = () => { cleanupProxy().catch(() => {}) }
+  process.once('SIGINT',  cleanup)
+  process.once('SIGTERM', cleanup)
+  process.once('exit',    cleanup)
+}
+
+export function buildProxyTopologyFromConfig(fcmConfig) {
+  const accounts = []
+  const proxyModels = {}
+
+  for (const merged of mergedModelsRef) {
+    proxyModels[merged.slug] = { name: merged.label }
+
+    for (const providerEntry of merged.providers) {
+      const keys = resolveApiKeys(fcmConfig, providerEntry.providerKey)
+      const providerSource = sources[providerEntry.providerKey]
+      if (!providerSource) continue
+
+      const rawUrl = resolveCloudflareUrl(providerSource.url)
+      const baseUrl = rawUrl.replace(/\/chat\/completions$/, '')
+
+      keys.forEach((apiKey, keyIdx) => {
+        accounts.push({
+          id: `${providerEntry.providerKey}/${merged.slug}/${keyIdx}`,
+          providerKey: providerEntry.providerKey,
+          proxyModelId: merged.slug,
+          modelId: providerEntry.modelId,
+          url: baseUrl,
+          apiKey,
+        })
+      })
+    }
+  }
+
+  return { accounts, proxyModels }
+}
+
+export async function ensureProxyRunning(fcmConfig, { forceRestart = false } = {}) {
+  registerExitHandlers()
+  proxyCleanedUp = false
+
+  if (forceRestart && activeProxy) {
+    await cleanupProxy()
+  }
+
+  const existingStatus = activeProxy?.getStatus?.()
+  if (existingStatus?.running === true) {
+    const availableModelSlugs = new Set(
+      (activeProxy._accounts || []).map(a => a.proxyModelId).filter(Boolean)
+    )
+    return {
+      port: existingStatus.port,
+      accountCount: existingStatus.accountCount,
+      proxyToken: activeProxy?._proxyApiKey,
+      proxyModels: null,
+      availableModelSlugs,
+    }
+  }
+
+  const { accounts, proxyModels } = buildProxyTopologyFromConfig(fcmConfig)
+  if (accounts.length === 0) {
+    throw new Error('No API keys found for proxy-capable models')
+  }
+
+  const proxyToken = `fcm_${randomUUID().replace(/-/g, '')}`
+  const proxy = new ProxyServer({ accounts, proxyApiKey: proxyToken })
+  const { port } = await proxy.start()
+  activeProxy = proxy
+  setActiveProxy(activeProxy)
+
+  const availableModelSlugs = new Set(accounts.map(a => a.proxyModelId).filter(Boolean))
+  return { port, accountCount: accounts.length, proxyToken, proxyModels, availableModelSlugs }
+}
+
+export async function autoStartProxyIfSynced(fcmConfig, state) {
+  try {
+    const ocConfig = loadOpenCodeConfig()
+    if (!ocConfig?.provider?.['fcm-proxy']) return
+
+    state.proxyStartupStatus = { phase: 'starting' }
+
+    const started = await ensureProxyRunning(fcmConfig)
+
+    syncToOpenCode(fcmConfig, sources, mergedModelsRef, {
+      proxyPort: started.port,
+      proxyToken: started.proxyToken,
+      availableModelSlugs: started.availableModelSlugs,
+    })
+
+    state.proxyStartupStatus = {
+      phase: 'running',
+      port: started.port,
+      accountCount: started.accountCount,
+    }
+  } catch (err) {
+    state.proxyStartupStatus = {
+      phase: 'failed',
+      reason: err?.message ?? String(err),
+    }
+  }
+}
+
+export async function startProxyAndLaunch(model, fcmConfig) {
+  try {
+    const started = await ensureProxyRunning(fcmConfig, { forceRestart: true })
+    const merged = mergedModelByLabelRef.get(model.label)
+    const defaultProxyModelId = merged?.slug ?? model.modelId
+
+    if (!started.proxyModels || Object.keys(started.proxyModels).length === 0) {
+      throw new Error('Proxy model catalog is empty')
+    }
+
+    console.log(chalk.dim(`  🔀 Multi-account proxy listening on port ${started.port} (${started.accountCount} accounts)`))
+    await startOpenCodeWithProxy(model, started.port, defaultProxyModelId, started.proxyModels, fcmConfig, started.proxyToken)
+  } catch (err) {
+    console.error(chalk.red(`  ✗ Proxy failed to start: ${err.message}`))
+    console.log(chalk.dim('  Falling back to direct single-account flow…'))
+    await cleanupProxy()
+    await startOpenCode(model, fcmConfig)
+  }
+}
+
+async function startOpenCodeWithProxy(model, port, proxyModelId, proxyModels, fcmConfig, proxyToken) {
+  const config = loadOpenCodeConfig()
+  if (!config.provider) config.provider = {}
+  const previousProxyProvider = config.provider['fcm-proxy']
+  const previousModel = config.model
+
+  const fallbackModelId = Object.keys(proxyModels)[0]
+  const selectedProxyModelId = proxyModels[proxyModelId] ? proxyModelId : fallbackModelId
+
+  config.provider['fcm-proxy'] = {
+    npm: '@ai-sdk/openai-compatible',
+    name: 'FCM Proxy',
+    options: {
+      baseURL: `http://127.0.0.1:${port}/v1`,
+      apiKey: proxyToken
+    },
+    models: proxyModels
+  }
+  config.model = `fcm-proxy/${selectedProxyModelId}`
+  saveOpenCodeConfig(config)
+
+  console.log(chalk.green(`  Setting ${chalk.bold(model.label)} via proxy as default for OpenCode…`))
+  console.log(chalk.dim(`  Model: fcm-proxy/${selectedProxyModelId}  •  Proxy: http://127.0.0.1:${port}/v1`))
+  console.log(chalk.dim(`  Catalog: ${Object.keys(proxyModels).length} models available via fcm-proxy`))
+  console.log()
+
+  try {
+    await spawnOpenCode(['--model', `fcm-proxy/${selectedProxyModelId}`], 'fcm-proxy', fcmConfig)
+  } finally {
+    try {
+      const savedCfg = loadOpenCodeConfig()
+      if (!savedCfg.provider) savedCfg.provider = {}
+
+      if (previousProxyProvider) {
+        savedCfg.provider['fcm-proxy'] = previousProxyProvider
+      } else if (savedCfg.provider['fcm-proxy']) {
+        delete savedCfg.provider['fcm-proxy']
+      }
+
+      if (typeof previousModel === 'string' && previousModel.length > 0) {
+        savedCfg.model = previousModel
+      } else if (typeof savedCfg.model === 'string' && savedCfg.model.startsWith('fcm-proxy/')) {
+        delete savedCfg.model
+      }
+
+      saveOpenCodeConfig(savedCfg)
+    } catch { /* best-effort */ }
+    await cleanupProxy()
+  }
+}
+
+// ─── Start OpenCode Desktop ───────────────────────────────────────────────────
+
+export async function startOpenCodeDesktop(model, fcmConfig) {
+  const providerKey = model.providerKey ?? 'nvidia'
+  const ocModelId = getOpenCodeModelId(providerKey, model.modelId)
+  const modelRef = `${providerKey}/${ocModelId}`
+
+  const launchDesktop = async () => {
+    const { exec } = await import('child_process')
+    let command
+    if (isMac) {
+      command = 'open -a OpenCode'
+    } else if (isWindows) {
+      command = 'start "" "%LOCALAPPDATA%\\Programs\\OpenCode\\OpenCode.exe" 2>nul || start "" "%PROGRAMFILES%\\OpenCode\\OpenCode.exe" 2>nul || start OpenCode'
+    } else if (isLinux) {
+      command = `opencode-desktop --model ${modelRef} 2>/dev/null || flatpak run ai.opencode.OpenCode --model ${modelRef} 2>/dev/null || snap run opencode --model ${modelRef} 2>/dev/null || xdg-open /usr/share/applications/opencode.desktop 2>/dev/null || echo "OpenCode not found"`
+    }
+    exec(command, (err) => {
+      if (err) {
+        console.error(chalk.red('  Could not open OpenCode Desktop'))
+        if (isWindows) {
+          console.error(chalk.dim('    Make sure OpenCode is installed from https://opencode.ai'))
+        } else if (isLinux) {
+          console.error(chalk.dim('    Install via: snap install opencode OR flatpak install ai.opencode.OpenCode'))
+          console.error(chalk.dim('    Or download from https://opencode.ai'))
+        } else {
+          console.error(chalk.dim('    Is it installed at /Applications/OpenCode.app?'))
+        }
+      }
+    })
+  }
+
+  if (providerKey === 'nvidia') {
+    const config = loadOpenCodeConfig()
+    const backupPath = `${getOpenCodeConfigPath()}.backup-${Date.now()}`
+
+    if (existsSync(getOpenCodeConfigPath())) {
+      copyFileSync(getOpenCodeConfigPath(), backupPath)
+      console.log(chalk.dim(`  Backup: ${backupPath}`))
+    }
+
+    if (!config.provider) config.provider = {}
+    if (!config.provider.nvidia) {
+      config.provider.nvidia = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'NVIDIA NIM',
+        options: {
+          baseURL: 'https://integrate.api.nvidia.com/v1',
+          apiKey: '{env:NVIDIA_API_KEY}'
+        },
+        models: {}
+      }
+      console.log(chalk.green('  + Auto-configured NVIDIA NIM provider in OpenCode'))
+    }
+
+    console.log(chalk.green(`  Setting ${chalk.bold(model.label)} as default for OpenCode Desktop...`))
+    console.log(chalk.dim(`  Model: ${modelRef}`))
+    console.log()
+
+    config.model = modelRef
+    if (!config.provider.nvidia.models) config.provider.nvidia.models = {}
+    config.provider.nvidia.models[ocModelId] = { name: model.label }
+
+    saveOpenCodeConfig(config)
+
+    const savedConfig = loadOpenCodeConfig()
+    console.log(chalk.dim(`  Config saved to: ${getOpenCodeConfigPath()}`))
+    console.log(chalk.dim(`  Default model in config: ${savedConfig.model || 'NOT SET'}`))
+    console.log()
+
+    if (savedConfig.model === config.model) {
+      console.log(chalk.green(`  Default model set to: ${modelRef}`))
+    } else {
+      console.log(chalk.yellow(`  Config might not have been saved correctly`))
+    }
+    console.log()
+    console.log(chalk.dim('  Opening OpenCode Desktop...'))
+    console.log()
+
+    await launchDesktop()
+    return
+  }
+
+  if (providerKey === 'replicate') {
+    console.log(chalk.yellow('  Replicate models are monitor-only for now in OpenCode Desktop mode.'))
+    console.log(chalk.dim('    Reason: Replicate uses /v1/predictions instead of OpenAI chat-completions.'))
+    console.log(chalk.dim('    You can still benchmark this model in the TUI and use other providers for Desktop launch.'))
+    console.log()
+    return
+  }
+
+  if (providerKey === 'zai') {
+    console.log(chalk.yellow('  ZAI models are supported in OpenCode CLI mode only (not Desktop).'))
+    console.log(chalk.dim('    Reason: ZAI requires a localhost proxy that only works with the CLI spawn.'))
+    console.log(chalk.dim('    Use OpenCode CLI mode (default) to launch ZAI models.'))
+    console.log()
+    return
+  }
+
+  console.log(chalk.green(`  Setting ${chalk.bold(model.label)} as default for OpenCode Desktop...`))
+  console.log(chalk.dim(`  Model: ${modelRef}`))
+  console.log()
+
+  const config = loadOpenCodeConfig()
+  const backupPath = `${getOpenCodeConfigPath()}.backup-${Date.now()}`
+
+  if (existsSync(getOpenCodeConfigPath())) {
+    copyFileSync(getOpenCodeConfigPath(), backupPath)
+    console.log(chalk.dim(`  Backup: ${backupPath}`))
+  }
+
+  if (!config.provider) config.provider = {}
+  if (!config.provider[providerKey]) {
+    if (providerKey === 'groq') {
+      config.provider.groq = { options: { apiKey: '{env:GROQ_API_KEY}' }, models: {} }
+    } else if (providerKey === 'cerebras') {
+      config.provider.cerebras = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Cerebras',
+        options: { baseURL: 'https://api.cerebras.ai/v1', apiKey: '{env:CEREBRAS_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'sambanova') {
+      config.provider.sambanova = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'SambaNova',
+        options: { baseURL: 'https://api.sambanova.ai/v1', apiKey: '{env:SAMBANOVA_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'openrouter') {
+      config.provider.openrouter = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'OpenRouter',
+        options: { baseURL: 'https://openrouter.ai/api/v1', apiKey: '{env:OPENROUTER_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'huggingface') {
+      config.provider.huggingface = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Hugging Face Inference',
+        options: { baseURL: 'https://router.huggingface.co/v1', apiKey: '{env:HUGGINGFACE_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'deepinfra') {
+      config.provider.deepinfra = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'DeepInfra',
+        options: { baseURL: 'https://api.deepinfra.com/v1/openai', apiKey: '{env:DEEPINFRA_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'fireworks') {
+      config.provider.fireworks = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Fireworks AI',
+        options: { baseURL: 'https://api.fireworks.ai/inference/v1', apiKey: '{env:FIREWORKS_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'codestral') {
+      config.provider.codestral = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Mistral Codestral',
+        options: { baseURL: 'https://codestral.mistral.ai/v1', apiKey: '{env:CODESTRAL_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'hyperbolic') {
+      config.provider.hyperbolic = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Hyperbolic',
+        options: { baseURL: 'https://api.hyperbolic.xyz/v1', apiKey: '{env:HYPERBOLIC_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'scaleway') {
+      config.provider.scaleway = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Scaleway',
+        options: { baseURL: 'https://api.scaleway.ai/v1', apiKey: '{env:SCALEWAY_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'googleai') {
+      config.provider.googleai = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Google AI Studio',
+        options: { baseURL: 'https://generativelanguage.googleapis.com/v1beta/openai', apiKey: '{env:GOOGLE_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'siliconflow') {
+      config.provider.siliconflow = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'SiliconFlow',
+        options: { baseURL: 'https://api.siliconflow.com/v1', apiKey: '{env:SILICONFLOW_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'together') {
+      config.provider.together = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Together AI',
+        options: { baseURL: 'https://api.together.xyz/v1', apiKey: '{env:TOGETHER_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'cloudflare') {
+      const cloudflareAccountId = (process.env.CLOUDFLARE_ACCOUNT_ID || '').trim()
+      if (!cloudflareAccountId) {
+        console.log(chalk.yellow('  Cloudflare Workers AI requires CLOUDFLARE_ACCOUNT_ID for OpenCode integration.'))
+        console.log(chalk.dim('    Export CLOUDFLARE_ACCOUNT_ID and retry this selection.'))
+        console.log()
+        return
+      }
+      config.provider.cloudflare = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Cloudflare Workers AI',
+        options: { baseURL: `https://api.cloudflare.com/client/v4/accounts/${cloudflareAccountId}/ai/v1`, apiKey: '{env:CLOUDFLARE_API_TOKEN}' },
+        models: {}
+      }
+    } else if (providerKey === 'perplexity') {
+      config.provider.perplexity = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'Perplexity API',
+        options: { baseURL: 'https://api.perplexity.ai', apiKey: '{env:PERPLEXITY_API_KEY}' },
+        models: {}
+      }
+    } else if (providerKey === 'iflow') {
+      config.provider.iflow = {
+        npm: '@ai-sdk/openai-compatible',
+        name: 'iFlow',
+        options: { baseURL: 'https://apis.iflow.cn/v1', apiKey: '{env:IFLOW_API_KEY}' },
+        models: {}
+      }
+    }
+  }
+
+  const isBuiltinMapped = OPENCODE_MODEL_MAP[providerKey]?.[model.modelId]
+  if (!isBuiltinMapped) {
+    if (!config.provider[providerKey].models) config.provider[providerKey].models = {}
+    config.provider[providerKey].models[ocModelId] = { name: model.label }
+  }
+
+  config.model = modelRef
+  saveOpenCodeConfig(config)
+
+  const savedConfig = loadOpenCodeConfig()
+  console.log(chalk.dim(`  Config saved to: ${getOpenCodeConfigPath()}`))
+  console.log(chalk.dim(`  Default model in config: ${savedConfig.model || 'NOT SET'}`))
+  console.log()
+
+  if (savedConfig.model === config.model) {
+    console.log(chalk.green(`  Default model set to: ${modelRef}`))
+  } else {
+    console.log(chalk.yellow(`  Config might not have been saved correctly`))
+  }
+  console.log()
+  console.log(chalk.dim('  Opening OpenCode Desktop...'))
+  console.log()
+
+  await launchDesktop()
+}