recruiter: add boss auto browse/chat flows, webhook, and candidate tables

- Add recruiter-side automation core and run-core entry
- Extend sqlite-plugin with candidate info + contact logs
- Add UI routes/pages, IPC handlers, progress + log panel
- Document current status and plans under plan/

Made-with: Cursor

packages/utils/gpt-request.mjs

@@ -1,10 +1,20 @@
 import OpenAI from "openai";
 
+/**
+ * 调用 Chat Completions API，支持推理模型（thinking 参数）。
+ *
+ * @param {{ baseURL: string, apiKey: string, model: string, max_tokens?: number, temperature?: number, thinking?: { enabled?: boolean, budget?: number } }} config
+ * @param {Array<{ role: string, content: string }>} messages
+ */
 export async function completes(
   {
     baseURL,
     apiKey,
-    model
+    model,
+    max_tokens,
+    temperature,
+    thinking,
+    response_format
   },
   messages
 ) {
@@ -13,14 +23,49 @@ export async function completes(
     apiKey,
   });
 
-  const completion = await openai.chat.completions.create({
+  const isThinking = !!(thinking?.enabled && thinking?.budget)
+
+  // 推理模型开启 thinking 时，max_tokens 必须大于 thinking_budget，否则会因长度上限截断 JSON。
+  // 调用方若未显式传 max_tokens，按是否启用 thinking 给一个安全的默认值。
+  const resolvedMaxTokens =
+    typeof max_tokens === 'number'
+      ? max_tokens
+      : isThinking
+        ? 8192
+        : 1200
+
+  // temperature：推理模型启用 thinking 时建议 ≥0.5（部分 provider 限制），普通 JSON 输出用 0.1。
+  const resolvedTemperature =
+    typeof temperature === 'number'
+      ? temperature
+      : isThinking
+        ? 0.6
+        : 0.1
+
+  const createParams = {
     messages,
     model,
-    frequency_penalty: 0,
-    max_tokens: 100,
-    temperature: 0.1
-  });
+    max_tokens: resolvedMaxTokens,
+    temperature: resolvedTemperature,
+  }
 
-  console.log(completion.choices[0].message.content);
+  if (isThinking) {
+    // SiliconFlow / 火山方舟等兼容顶层参数；OpenAI SDK 通过 extra_body 透传其他字段
+    createParams.enable_thinking = true
+    createParams.thinking_budget = thinking.budget
+  }
+
+  if (response_format) {
+    createParams.response_format = response_format
+  }
+
+  const completion = await openai.chat.completions.create(createParams);
+
+  // reasoning_content 仅推理模型填充，普通模型为 undefined
+  const msg = completion.choices[0].message
+  if (msg.reasoning_content) {
+    console.log('[gpt-request] reasoning_content:', String(msg.reasoning_content ?? '').slice(0, 200))
+  }
+  console.log('[gpt-request] content:', (msg.content ?? '').slice(0, 200));
   return completion;
 }

packages/utils/sleep.mjs

@@ -4,6 +4,8 @@ export function sleep (t) {
   })
 }
 
-export function sleepWithRandomDelay (base) {
-  return sleep(base + Math.random()*1000)
+export function sleepWithRandomDelay (min, max) {
+  const lo = min ?? 0
+  const hi = max ?? (lo + 1000)
+  return sleep(lo + Math.random() * (hi - lo))
 }