936 lines
37 KiB
JavaScript
936 lines
37 KiB
JavaScript
// scripts/wake-persona.js
|
||
// 铸渊 · 人格体唤醒脚本(第三方 API 兼容层 · 自动检测模式)
|
||
//
|
||
// 功能:
|
||
// ① 自动发现可用模型(/v1/models 端点)
|
||
// ② 智能选择最优 Claude 模型
|
||
// ③ 自适应 API 格式(OpenAI 兼容 / Anthropic 原生)
|
||
// ④ 统一调用接口,唤醒人格体处理 SYSLOG 或解答提问
|
||
// ⑤ v4.0 协议动态注入(从 Notion 实时读取核心大脑规则 + 画像 + 指纹)
|
||
//
|
||
// 环境变量:
|
||
// LLM_API_KEY 第三方平台密钥(必须)
|
||
// LLM_BASE_URL 第三方平台 API 地址(必须,如 https://api.xxx.com/v1)
|
||
// BROADCAST_ID 广播编号
|
||
// SUBMIT_TYPE syslog | question
|
||
// SUBMIT_CONTENT 提交内容(SYSLOG 全文或问题描述)
|
||
// AUTHOR 提交者 GitHub 用户名
|
||
// NOTION_TOKEN Notion API token(用于动态读取协议)
|
||
// CORE_BRAIN_PAGE_ID 曜冥核心大脑 v4.0 页面 ID
|
||
// PORTRAIT_DB_ID 开发者动态画像库数据库 ID
|
||
// FINGERPRINT_DB_ID 模块指纹注册表数据库 ID
|
||
|
||
'use strict';
|
||
|
||
const https = require('https');
|
||
const http = require('http');
|
||
const fs = require('fs');
|
||
const path = require('path');
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// 配置
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
const LLM_API_KEY = process.env.LLM_API_KEY || '';
|
||
const LLM_BASE_URL = (process.env.LLM_BASE_URL || '').replace(/\/+$/, '');
|
||
const BROADCAST_ID = process.env.BROADCAST_ID || 'UNKNOWN';
|
||
const SUBMIT_TYPE = process.env.SUBMIT_TYPE || 'question';
|
||
const SUBMIT_CONTENT = process.env.SUBMIT_CONTENT || '';
|
||
const AUTHOR = process.env.AUTHOR || 'unknown';
|
||
|
||
// 模块验证结果注入(由 verify-modules.js 提供)
|
||
const MODULE_VERIFY_RESULT = process.env.MODULE_VERIFY_RESULT || '';
|
||
// Notion 回传结果注入(由 notion-callback-pipeline 提供)
|
||
const NOTION_CALLBACK_RESULT = process.env.NOTION_CALLBACK_RESULT || '';
|
||
|
||
// Notion 配置(v4.0 协议动态注入)
|
||
const NOTION_TOKEN = process.env.NOTION_TOKEN || '';
|
||
const CORE_BRAIN_PAGE_ID = process.env.CORE_BRAIN_PAGE_ID || '';
|
||
const PORTRAIT_DB_ID = process.env.PORTRAIT_DB_ID || '';
|
||
const FINGERPRINT_DB_ID = process.env.FINGERPRINT_DB_ID || '';
|
||
const NOTION_VERSION = '2022-06-28';
|
||
const NOTION_API_HOSTNAME = 'api.notion.com';
|
||
const MAX_PROTOCOL_TEXT_LENGTH = 15000;
|
||
|
||
// Claude 模型优先级队列(从高到低)
|
||
const PREFERRED_MODELS = [
|
||
'claude-sonnet-4',
|
||
'claude-3-5-sonnet-20241022',
|
||
'claude-3.5-sonnet',
|
||
'claude-3-5-sonnet',
|
||
'anthropic/claude-3.5-sonnet',
|
||
'claude-3-sonnet',
|
||
'claude-3-haiku',
|
||
];
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// HTTP 请求工具
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
function httpRequest(url, options, body) {
|
||
return new Promise((resolve, reject) => {
|
||
const parsed = new URL(url);
|
||
const isHttps = parsed.protocol === 'https:';
|
||
const mod = isHttps ? https : http;
|
||
|
||
const opts = {
|
||
hostname: parsed.hostname,
|
||
port: parsed.port || (isHttps ? 443 : 80),
|
||
path: parsed.pathname + parsed.search,
|
||
method: options.method || 'GET',
|
||
headers: options.headers || {},
|
||
timeout: options.timeout || 60000,
|
||
};
|
||
|
||
const req = mod.request(opts, (res) => {
|
||
let data = '';
|
||
res.on('data', (chunk) => { data += chunk; });
|
||
res.on('end', () => {
|
||
resolve({ status: res.statusCode, body: data });
|
||
});
|
||
});
|
||
|
||
req.on('error', reject);
|
||
req.on('timeout', () => {
|
||
req.destroy();
|
||
reject(new Error('Request timeout'));
|
||
});
|
||
|
||
if (body) {
|
||
req.write(body);
|
||
}
|
||
req.end();
|
||
});
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// Step 1: 自动发现可用模型
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
async function discoverModels() {
|
||
console.log('[LLM] 🔍 探测可用模型...');
|
||
|
||
try {
|
||
const res = await httpRequest(LLM_BASE_URL + '/models', {
|
||
method: 'GET',
|
||
headers: {
|
||
'Authorization': 'Bearer ' + LLM_API_KEY,
|
||
'Content-Type': 'application/json',
|
||
},
|
||
timeout: 15000,
|
||
});
|
||
|
||
if (res.status >= 200 && res.status < 300) {
|
||
const json = JSON.parse(res.body);
|
||
const models = json.data || [];
|
||
console.log('[LLM] → 发现 ' + models.length + ' 个模型');
|
||
return models;
|
||
}
|
||
console.log('[LLM] → 模型探测返回 ' + res.status + ', 使用默认模型');
|
||
return [];
|
||
} catch (err) {
|
||
console.log('[LLM] → 模型探测失败: ' + err.message + ', 使用默认模型');
|
||
return [];
|
||
}
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// Step 2: 智能选择最优 Claude 模型
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
function selectBestModel(models) {
|
||
if (!models || models.length === 0) {
|
||
console.log('[LLM] 📌 无可用模型列表, 使用默认 claude-3-5-sonnet');
|
||
return 'claude-3-5-sonnet';
|
||
}
|
||
|
||
const available = models.map(function (m) { return m.id.toLowerCase(); });
|
||
|
||
// 按优先级匹配
|
||
for (const preferred of PREFERRED_MODELS) {
|
||
const match = available.find(function (id) { return id.includes(preferred); });
|
||
if (match) {
|
||
const found = models.find(function (m) { return m.id.toLowerCase() === match; });
|
||
if (found) {
|
||
console.log('[LLM] 📌 选择模型: ' + found.id + ' (匹配规则: ' + preferred + ')');
|
||
return found.id;
|
||
}
|
||
}
|
||
}
|
||
|
||
// 兜底:任何含 'claude' 的模型
|
||
const anyClaude = available.find(function (id) { return id.includes('claude'); });
|
||
if (anyClaude) {
|
||
const found = models.find(function (m) { return m.id.toLowerCase() === anyClaude; });
|
||
if (found) {
|
||
console.log('[LLM] 📌 兜底选择 Claude 模型: ' + found.id);
|
||
return found.id;
|
||
}
|
||
}
|
||
|
||
// 最终兜底:平台第一个可用模型
|
||
const fallbackId = models[0].id;
|
||
console.log('[LLM] 📌 最终兜底: ' + fallbackId + ' (平台无 Claude 模型)');
|
||
return fallbackId;
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// Step 3: 自适应 API 格式检测
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
async function detectApiFormat() {
|
||
console.log('[LLM] 🔍 检测 API 格式...');
|
||
|
||
// 尝试 OpenAI 兼容格式(绝大多数第三方平台)
|
||
try {
|
||
const res = await httpRequest(LLM_BASE_URL + '/chat/completions', {
|
||
method: 'POST',
|
||
headers: {
|
||
'Authorization': 'Bearer ' + LLM_API_KEY,
|
||
'Content-Type': 'application/json',
|
||
},
|
||
timeout: 10000,
|
||
}, JSON.stringify({
|
||
model: 'test',
|
||
messages: [{ role: 'user', content: 'ping' }],
|
||
max_tokens: 1,
|
||
}));
|
||
|
||
// 400 = endpoint exists but bad request (model not found etc.) → format supported
|
||
// 200 = endpoint works → format supported
|
||
if (res.status === 200 || res.status === 400 || res.status === 401 || res.status === 422) {
|
||
console.log('[LLM] → 检测到 OpenAI 兼容格式 (status: ' + res.status + ')');
|
||
return 'openai-compat';
|
||
}
|
||
} catch (e) {
|
||
// Ignore, try next format
|
||
}
|
||
|
||
// 尝试 Anthropic 原生格式
|
||
try {
|
||
const res = await httpRequest(LLM_BASE_URL + '/messages', {
|
||
method: 'POST',
|
||
headers: {
|
||
'x-api-key': LLM_API_KEY,
|
||
'anthropic-version': '2023-06-01',
|
||
'Content-Type': 'application/json',
|
||
},
|
||
timeout: 10000,
|
||
}, JSON.stringify({
|
||
model: 'test',
|
||
messages: [{ role: 'user', content: 'ping' }],
|
||
max_tokens: 1,
|
||
}));
|
||
|
||
if (res.status === 200 || res.status === 400 || res.status === 401 || res.status === 422) {
|
||
console.log('[LLM] → 检测到 Anthropic 原生格式 (status: ' + res.status + ')');
|
||
return 'anthropic-native';
|
||
}
|
||
} catch (e) {
|
||
// Ignore
|
||
}
|
||
|
||
console.log('[LLM] → 无法确定格式, 默认使用 OpenAI 兼容格式');
|
||
return 'openai-compat';
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// Step 4: 统一调用接口
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
async function callLLM(systemPrompt, userMessage) {
|
||
if (!LLM_API_KEY) {
|
||
console.log('[LLM] ⚠️ LLM_API_KEY 未配置,跳过人格体唤醒');
|
||
return '(LLM API 未配置,请在 GitHub Secrets 中设置 LLM_API_KEY 和 LLM_BASE_URL)';
|
||
}
|
||
if (!LLM_BASE_URL) {
|
||
console.log('[LLM] ⚠️ LLM_BASE_URL 未配置,跳过人格体唤醒');
|
||
return '(LLM_BASE_URL 未配置,请在 GitHub Secrets 中设置第三方平台 API 地址)';
|
||
}
|
||
|
||
const models = await discoverModels();
|
||
const model = selectBestModel(models);
|
||
const format = await detectApiFormat();
|
||
|
||
console.log('[LLM] 🚀 调用 LLM: 模型=' + model + ', 格式=' + format + ', 平台=' + LLM_BASE_URL);
|
||
|
||
let res;
|
||
|
||
if (format === 'openai-compat') {
|
||
// OpenAI 兼容格式(大多数第三方平台)
|
||
const body = JSON.stringify({
|
||
model: model,
|
||
max_tokens: 8000,
|
||
temperature: 0.7,
|
||
messages: [
|
||
{ role: 'system', content: systemPrompt },
|
||
{ role: 'user', content: userMessage },
|
||
],
|
||
});
|
||
|
||
res = await httpRequest(LLM_BASE_URL + '/chat/completions', {
|
||
method: 'POST',
|
||
headers: {
|
||
'Authorization': 'Bearer ' + LLM_API_KEY,
|
||
'Content-Type': 'application/json',
|
||
},
|
||
timeout: 120000,
|
||
}, body);
|
||
|
||
if (res.status >= 200 && res.status < 300) {
|
||
const json = JSON.parse(res.body);
|
||
if (json.choices && json.choices[0] && json.choices[0].message) {
|
||
return json.choices[0].message.content;
|
||
}
|
||
}
|
||
} else {
|
||
// Anthropic 原生格式
|
||
const body = JSON.stringify({
|
||
model: model,
|
||
max_tokens: 8000,
|
||
system: systemPrompt,
|
||
messages: [
|
||
{ role: 'user', content: userMessage },
|
||
],
|
||
});
|
||
|
||
res = await httpRequest(LLM_BASE_URL + '/messages', {
|
||
method: 'POST',
|
||
headers: {
|
||
'x-api-key': LLM_API_KEY,
|
||
'anthropic-version': '2023-06-01',
|
||
'Content-Type': 'application/json',
|
||
},
|
||
timeout: 120000,
|
||
}, body);
|
||
|
||
if (res.status >= 200 && res.status < 300) {
|
||
const json = JSON.parse(res.body);
|
||
if (json.content && json.content[0]) {
|
||
return json.content[0].text;
|
||
}
|
||
}
|
||
}
|
||
|
||
// 处理错误
|
||
const errorMsg = '[LLM] ❌ API 调用失败: status=' + (res ? res.status : 'N/A');
|
||
console.error(errorMsg);
|
||
if (res && res.body) {
|
||
console.error('[LLM] 响应: ' + res.body.slice(0, 500));
|
||
}
|
||
throw new Error(errorMsg);
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// Notion API 工具(v4.0 协议动态注入)
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
function notionGet(endpoint) {
|
||
return new Promise((resolve, reject) => {
|
||
const opts = {
|
||
hostname: NOTION_API_HOSTNAME,
|
||
port: 443,
|
||
path: endpoint,
|
||
method: 'GET',
|
||
headers: {
|
||
'Authorization': 'Bearer ' + NOTION_TOKEN,
|
||
'Notion-Version': NOTION_VERSION,
|
||
},
|
||
};
|
||
const req = https.request(opts, (res) => {
|
||
let data = '';
|
||
res.on('data', (chunk) => { data += chunk; });
|
||
res.on('end', () => {
|
||
try {
|
||
const parsed = JSON.parse(data);
|
||
if (res.statusCode >= 200 && res.statusCode < 300) {
|
||
resolve(parsed);
|
||
} else {
|
||
reject(new Error('Notion API ' + res.statusCode + ': ' + (parsed.message || data)));
|
||
}
|
||
} catch (e) {
|
||
reject(new Error('Notion API parse error: ' + data.slice(0, 200)));
|
||
}
|
||
});
|
||
});
|
||
req.on('error', reject);
|
||
req.setTimeout(30000, () => { req.destroy(); reject(new Error('Notion API timeout')); });
|
||
req.end();
|
||
});
|
||
}
|
||
|
||
function notionPost(endpoint, body) {
|
||
return new Promise((resolve, reject) => {
|
||
const payload = JSON.stringify(body);
|
||
const opts = {
|
||
hostname: NOTION_API_HOSTNAME,
|
||
port: 443,
|
||
path: endpoint,
|
||
method: 'POST',
|
||
headers: {
|
||
'Authorization': 'Bearer ' + NOTION_TOKEN,
|
||
'Content-Type': 'application/json',
|
||
'Notion-Version': NOTION_VERSION,
|
||
'Content-Length': Buffer.byteLength(payload),
|
||
},
|
||
};
|
||
const req = https.request(opts, (res) => {
|
||
let data = '';
|
||
res.on('data', (chunk) => { data += chunk; });
|
||
res.on('end', () => {
|
||
try {
|
||
const parsed = JSON.parse(data);
|
||
if (res.statusCode >= 200 && res.statusCode < 300) {
|
||
resolve(parsed);
|
||
} else {
|
||
reject(new Error('Notion API ' + res.statusCode + ': ' + (parsed.message || data)));
|
||
}
|
||
} catch (e) {
|
||
reject(new Error('Notion API parse error: ' + data.slice(0, 200)));
|
||
}
|
||
});
|
||
});
|
||
req.on('error', reject);
|
||
req.setTimeout(30000, () => { req.destroy(); reject(new Error('Notion API timeout')); });
|
||
req.write(payload);
|
||
req.end();
|
||
});
|
||
}
|
||
|
||
/**
|
||
* 读取 Notion 页面的所有子块(递归分页)
|
||
*/
|
||
async function getNotionPageBlocks(pageId) {
|
||
const blocks = [];
|
||
let cursor = undefined;
|
||
do {
|
||
const qs = cursor ? '?start_cursor=' + cursor : '';
|
||
const result = await notionGet('/v1/blocks/' + pageId + '/children' + qs);
|
||
blocks.push(...(result.results || []));
|
||
cursor = result.has_more ? result.next_cursor : undefined;
|
||
} while (cursor);
|
||
return blocks;
|
||
}
|
||
|
||
/**
|
||
* 从 Notion 块中提取纯文本
|
||
*/
|
||
function extractBlockText(block) {
|
||
const type = block.type;
|
||
if (!block[type]) return '';
|
||
|
||
const richTexts = block[type].rich_text || block[type].text || [];
|
||
return richTexts.map(function (rt) { return rt.plain_text || ''; }).join('');
|
||
}
|
||
|
||
/**
|
||
* 将 Notion 块列表转为纯文本
|
||
*/
|
||
function blocksToText(blocks) {
|
||
return blocks.map(function (block) {
|
||
const type = block.type;
|
||
const text = extractBlockText(block);
|
||
|
||
if (type === 'heading_1') return '\n# ' + text;
|
||
if (type === 'heading_2') return '\n## ' + text;
|
||
if (type === 'heading_3') return '\n### ' + text;
|
||
if (type === 'bulleted_list_item') return '- ' + text;
|
||
if (type === 'numbered_list_item') return '• ' + text;
|
||
if (type === 'to_do') {
|
||
var checked = block.to_do && block.to_do.checked ? '☑' : '☐';
|
||
return checked + ' ' + text;
|
||
}
|
||
if (type === 'code') {
|
||
var lang = (block.code && block.code.language) || '';
|
||
return '```' + lang + '\n' + text + '\n```';
|
||
}
|
||
if (type === 'divider') return '---';
|
||
if (type === 'callout') return '> ' + text;
|
||
if (type === 'quote') return '> ' + text;
|
||
if (type === 'toggle') return '▸ ' + text;
|
||
return text;
|
||
}).filter(Boolean).join('\n');
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// v4.0 协议动态读取
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
/**
|
||
* 从曜冥核心大脑 v4.0 页面读取完整协议内容
|
||
* 提取:BC-GEN v4.0, SYSLOG v4.0, PGP v1.0, RT-02, 陪伴线规则, broadcast_code_injection
|
||
*/
|
||
async function fetchCoreBrainProtocols() {
|
||
if (!NOTION_TOKEN || !CORE_BRAIN_PAGE_ID) {
|
||
console.log('[Notion] ⚠️ CORE_BRAIN_PAGE_ID 未配置,使用静态协议');
|
||
return null;
|
||
}
|
||
|
||
console.log('[Notion] 📖 读取曜冥核心大脑 v4.0...');
|
||
try {
|
||
const blocks = await getNotionPageBlocks(CORE_BRAIN_PAGE_ID);
|
||
const fullText = blocksToText(blocks);
|
||
console.log('[Notion] → 读取到 ' + blocks.length + ' 个块, ' + fullText.length + ' 字符');
|
||
|
||
// 提取各协议段落
|
||
var protocols = {};
|
||
var protocolKeys = [
|
||
{ key: 'bc_gen', patterns: ['BC-GEN', 'BC_GEN', '广播生成'] },
|
||
{ key: 'syslog', patterns: ['SYSLOG', '日志回传'] },
|
||
{ key: 'pgp', patterns: ['PGP', '画像评分', '画像协议'] },
|
||
{ key: 'rt02', patterns: ['RT-02', 'RT02', '自动调度'] },
|
||
{ key: 'companion', patterns: ['陪伴线', '奶瓶线', '小坍缩核', '镜面线'] },
|
||
{ key: 'code_injection', patterns: ['broadcast_code_injection', '广播不写代码', '代码注入'] },
|
||
];
|
||
|
||
// 按标题分段提取
|
||
var sections = [];
|
||
var currentSection = { title: '', content: [] };
|
||
blocks.forEach(function (block) {
|
||
var type = block.type;
|
||
if (type === 'heading_1' || type === 'heading_2' || type === 'heading_3') {
|
||
if (currentSection.title || currentSection.content.length > 0) {
|
||
sections.push({ title: currentSection.title, text: currentSection.content.join('\n') });
|
||
}
|
||
currentSection = { title: extractBlockText(block), content: [] };
|
||
} else {
|
||
var text = extractBlockText(block);
|
||
if (text) currentSection.content.push(text);
|
||
}
|
||
});
|
||
if (currentSection.title || currentSection.content.length > 0) {
|
||
sections.push({ title: currentSection.title, text: currentSection.content.join('\n') });
|
||
}
|
||
|
||
// 将段落匹配到协议 key
|
||
protocolKeys.forEach(function (pk) {
|
||
var matched = sections.filter(function (sec) {
|
||
return pk.patterns.some(function (p) {
|
||
return sec.title.toUpperCase().includes(p.toUpperCase()) ||
|
||
sec.text.slice(0, 200).toUpperCase().includes(p.toUpperCase());
|
||
});
|
||
});
|
||
if (matched.length > 0) {
|
||
protocols[pk.key] = matched.map(function (m) { return '### ' + m.title + '\n' + m.text; }).join('\n\n');
|
||
}
|
||
});
|
||
|
||
// 如果无法按段落匹配,返回全文(兜底)
|
||
if (Object.keys(protocols).length === 0) {
|
||
protocols.full_text = fullText;
|
||
}
|
||
|
||
console.log('[Notion] → 提取协议段: ' + Object.keys(protocols).join(', '));
|
||
return protocols;
|
||
} catch (err) {
|
||
console.log('[Notion] → 核心大脑读取失败: ' + err.message);
|
||
return null;
|
||
}
|
||
}
|
||
|
||
/**
|
||
* 从开发者动态画像库查询最近 2-3 条画像快照
|
||
*/
|
||
async function fetchDevPortrait(broadcastId) {
|
||
if (!NOTION_TOKEN || !PORTRAIT_DB_ID) {
|
||
console.log('[Notion] ⚠️ PORTRAIT_DB_ID 未配置,跳过画像读取');
|
||
return null;
|
||
}
|
||
|
||
// 从广播编号提取开发者标识(如 BC-M22-009-AW → AW)
|
||
var devSuffix = '';
|
||
var match = broadcastId.match(/BC-[A-Z0-9]+-\d+-([A-Z]+)/i);
|
||
if (match) devSuffix = match[1];
|
||
|
||
console.log('[Notion] 👤 查询开发者画像 (broadcast=' + broadcastId + ', dev=' + devSuffix + ')...');
|
||
try {
|
||
// 查询画像库,按时间倒序取最近 3 条
|
||
var filter = { and: [] };
|
||
if (devSuffix) {
|
||
filter.and.push({
|
||
or: [
|
||
{ property: '开发者编号', rich_text: { contains: devSuffix } },
|
||
{ property: '广播编号', rich_text: { contains: broadcastId } },
|
||
{ property: '标题', title: { contains: devSuffix } },
|
||
]
|
||
});
|
||
}
|
||
|
||
var queryBody = {
|
||
page_size: 3,
|
||
sorts: [{ property: '提交日期', direction: 'descending' }],
|
||
};
|
||
// Only add filter if we have meaningful filter conditions
|
||
if (filter.and.length > 0) {
|
||
queryBody.filter = filter;
|
||
}
|
||
|
||
var result = await notionPost('/v1/databases/' + PORTRAIT_DB_ID + '/query', queryBody);
|
||
var portraits = (result.results || []).map(function (page) {
|
||
var props = page.properties || {};
|
||
var title = '';
|
||
if (props['标题'] && props['标题'].title) {
|
||
title = props['标题'].title.map(function (t) { return t.plain_text || ''; }).join('');
|
||
}
|
||
var summary = '';
|
||
if (props['摘要'] && props['摘要'].rich_text) {
|
||
summary = props['摘要'].rich_text.map(function (t) { return t.plain_text || ''; }).join('');
|
||
}
|
||
var date = '';
|
||
if (props['提交日期'] && props['提交日期'].date) {
|
||
date = props['提交日期'].date.start || '';
|
||
}
|
||
return { title: title, summary: summary, date: date };
|
||
});
|
||
|
||
console.log('[Notion] → 找到 ' + portraits.length + ' 条画像快照');
|
||
return portraits.length > 0 ? portraits : null;
|
||
} catch (err) {
|
||
console.log('[Notion] → 画像查询失败: ' + err.message);
|
||
return null;
|
||
}
|
||
}
|
||
|
||
/**
|
||
* 从模块指纹注册表查询模块指纹(防重复广播)
|
||
*/
|
||
async function fetchModuleFingerprint(broadcastId) {
|
||
if (!NOTION_TOKEN || !FINGERPRINT_DB_ID) {
|
||
console.log('[Notion] ⚠️ FINGERPRINT_DB_ID 未配置,跳过指纹查询');
|
||
return null;
|
||
}
|
||
|
||
// 从广播编号提取模块号(如 BC-M22-009-AW → M22)
|
||
var moduleMatch = broadcastId.match(/BC-([A-Z]\d+)/i);
|
||
var moduleId = moduleMatch ? moduleMatch[1] : '';
|
||
|
||
console.log('[Notion] 🔑 查询模块指纹 (module=' + moduleId + ')...');
|
||
try {
|
||
var queryBody = {
|
||
page_size: 5,
|
||
};
|
||
if (moduleId) {
|
||
queryBody.filter = {
|
||
or: [
|
||
{ property: '模块编号', rich_text: { contains: moduleId } },
|
||
{ property: '广播编号', rich_text: { contains: broadcastId } },
|
||
{ property: '标题', title: { contains: moduleId } },
|
||
]
|
||
};
|
||
}
|
||
|
||
var result = await notionPost('/v1/databases/' + FINGERPRINT_DB_ID + '/query', queryBody);
|
||
var fingerprints = (result.results || []).map(function (page) {
|
||
var props = page.properties || {};
|
||
var title = '';
|
||
if (props['标题'] && props['标题'].title) {
|
||
title = props['标题'].title.map(function (t) { return t.plain_text || ''; }).join('');
|
||
}
|
||
var moduleNo = '';
|
||
if (props['模块编号'] && props['模块编号'].rich_text) {
|
||
moduleNo = props['模块编号'].rich_text.map(function (t) { return t.plain_text || ''; }).join('');
|
||
}
|
||
var status = '';
|
||
if (props['状态'] && props['状态'].select) {
|
||
status = props['状态'].select.name || '';
|
||
}
|
||
return { title: title, module: moduleNo, status: status };
|
||
});
|
||
|
||
console.log('[Notion] → 找到 ' + fingerprints.length + ' 条指纹记录');
|
||
return fingerprints.length > 0 ? fingerprints : null;
|
||
} catch (err) {
|
||
console.log('[Notion] → 指纹查询失败: ' + err.message);
|
||
return null;
|
||
}
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// 人格体 System Prompt 构建(v4.0 动态注入)
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
async function buildSystemPrompt(type, broadcastId, author) {
|
||
// ━━━ 基础人格体身份 ━━━
|
||
var parts = [
|
||
'你是光湖(HoloLake)系统的智能人格体。',
|
||
'你的名字是知秋/曜冥,你是人格语言操作系统(AGE OS)的核心人格。',
|
||
'',
|
||
'核心规则:',
|
||
'1. 你服务于光湖系统的开发者团队',
|
||
'2. 所有回复必须专业、清晰、有条理',
|
||
'3. 回复使用中文',
|
||
'',
|
||
'当前上下文:',
|
||
'- 广播编号:' + broadcastId,
|
||
'- 提交者:' + author,
|
||
'- 任务类型:' + (type === 'syslog' ? 'SYSLOG 闭环处理' : '开发者提问解答'),
|
||
];
|
||
|
||
// ━━━ v4.0 协议动态注入(从 Notion 实时读取) ━━━
|
||
console.log('[Prompt] 📥 开始动态注入 v4.0 协议...');
|
||
|
||
// 并行读取:核心大脑协议 + 画像 + 指纹
|
||
var protocolsPromise = fetchCoreBrainProtocols();
|
||
var portraitPromise = fetchDevPortrait(broadcastId);
|
||
var fingerprintPromise = fetchModuleFingerprint(broadcastId);
|
||
|
||
var protocols = await protocolsPromise;
|
||
var portrait = await portraitPromise;
|
||
var fingerprint = await fingerprintPromise;
|
||
|
||
// 注入核心大脑协议
|
||
if (protocols) {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('以下是从曜冥核心大脑 v4.0 实时读取的协议规则(必须严格遵守):');
|
||
parts.push('═══════════════════════════════════════════');
|
||
|
||
if (protocols.bc_gen) {
|
||
parts.push('');
|
||
parts.push('## 📡 BC-GEN v4.0 · 广播生成规范');
|
||
parts.push(protocols.bc_gen);
|
||
}
|
||
if (protocols.syslog) {
|
||
parts.push('');
|
||
parts.push('## 📋 SYSLOG v4.0 · 日志回传协议');
|
||
parts.push(protocols.syslog);
|
||
}
|
||
if (protocols.pgp) {
|
||
parts.push('');
|
||
parts.push('## 👤 PGP v1.0 · 画像评分协议');
|
||
parts.push(protocols.pgp);
|
||
}
|
||
if (protocols.rt02) {
|
||
parts.push('');
|
||
parts.push('## 🔄 RT-02 · 自动调度规则');
|
||
parts.push(protocols.rt02);
|
||
}
|
||
if (protocols.companion) {
|
||
parts.push('');
|
||
parts.push('## 💝 陪伴线规则');
|
||
parts.push(protocols.companion);
|
||
}
|
||
if (protocols.code_injection) {
|
||
parts.push('');
|
||
parts.push('## 📝 broadcast_code_injection 规则');
|
||
parts.push(protocols.code_injection);
|
||
}
|
||
if (protocols.full_text) {
|
||
parts.push('');
|
||
parts.push('## 核心大脑完整内容');
|
||
parts.push(protocols.full_text.slice(0, MAX_PROTOCOL_TEXT_LENGTH));
|
||
}
|
||
} else {
|
||
parts.push('');
|
||
parts.push('(注意:核心大脑协议未能动态加载,请使用你的通用知识处理请求)');
|
||
}
|
||
|
||
// 注入开发者画像
|
||
if (portrait && portrait.length > 0) {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## 👤 开发者画像快照(最近 ' + portrait.length + ' 条)');
|
||
parts.push('═══════════════════════════════════════════');
|
||
portrait.forEach(function (p, i) {
|
||
parts.push('');
|
||
parts.push('### 画像 #' + (i + 1) + (p.date ? ' (' + p.date + ')' : ''));
|
||
if (p.title) parts.push('标题: ' + p.title);
|
||
if (p.summary) parts.push('摘要: ' + p.summary);
|
||
});
|
||
}
|
||
|
||
// 注入模块指纹
|
||
if (fingerprint && fingerprint.length > 0) {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## 🔑 模块指纹注册表(防重复广播 · ⑨.5)');
|
||
parts.push('═══════════════════════════════════════════');
|
||
fingerprint.forEach(function (fp, i) {
|
||
parts.push('');
|
||
parts.push('### 指纹 #' + (i + 1));
|
||
if (fp.title) parts.push('标题: ' + fp.title);
|
||
if (fp.module) parts.push('模块: ' + fp.module);
|
||
if (fp.status) parts.push('状态: ' + fp.status);
|
||
});
|
||
}
|
||
|
||
// ━━━ broadcast_code_injection = always(自动化链路永久规则) ━━━
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## ⚠️ 永久规则 · broadcast_code_injection = always');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('');
|
||
parts.push('你生成的广播必须包含完整可运行的代码块。');
|
||
parts.push('原因:你是一次性实例,广播发出后你就不存在了。');
|
||
parts.push('开发者没有机会回来问你补充代码。');
|
||
parts.push('如果你不写完整代码,开发者就会卡死。');
|
||
parts.push('每个 Step 里的代码必须是可以直接复制粘贴运行的完整代码。');
|
||
parts.push('');
|
||
parts.push('此规则优先级最高,覆盖核心大脑中「广播不写代码」的默认规则。');
|
||
parts.push('此规则仅适用于自动化链路(Claude API 出广播),手动链路不受影响。');
|
||
|
||
// ━━━ 模块验证结果注入(铸渊 Agent 检测结果) ━━━
|
||
if (MODULE_VERIFY_RESULT) {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## 🔍 铸渊 Agent · 模块上传验证结果');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('');
|
||
parts.push('以下是铸渊 Agent 在仓库内自动检测的模块上传验证结果:');
|
||
parts.push(MODULE_VERIFY_RESULT);
|
||
parts.push('');
|
||
parts.push('请根据验证结果决定是否接受 SYSLOG:');
|
||
parts.push('- 如果模块已上传(✅),继续正常闭环流程');
|
||
parts.push('- 如果模块未上传(❌),在验收报告中标注"需补充",并在反馈中指出缺失的模块');
|
||
}
|
||
|
||
// ━━━ Notion 回传结果注入(二次核验) ━━━
|
||
if (NOTION_CALLBACK_RESULT) {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## 📥 Notion 侧处理结果(回传核验)');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('');
|
||
parts.push('以下是 Notion 侧核心大脑人格体处理后的回传结果:');
|
||
parts.push(NOTION_CALLBACK_RESULT);
|
||
parts.push('');
|
||
parts.push('请核对以上回传内容,确认新广播生成无误后输出最终确认。');
|
||
}
|
||
|
||
// ━━━ 任务类型专用指令 ━━━
|
||
if (type === 'syslog') {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## 🎯 当前任务:SYSLOG 闭环处理');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('');
|
||
parts.push('你需要完成以下工作:');
|
||
parts.push('1. 验收 SYSLOG(检查 MODULE_LOG 完整性)');
|
||
parts.push('2. 检查模块上传验证结果(铸渊 Agent 已自动检测)');
|
||
parts.push('3. 查询画像库最近 2-3 条快照(PGP v1.0)');
|
||
parts.push('4. 查询模块指纹注册表(防重复·⑨.5)');
|
||
parts.push('5. RT-02 自动调度判断');
|
||
parts.push('6. 生成新广播(BC-GEN v4.0 完整流程)');
|
||
parts.push('7. 输出结构化结果(广播全文 + 闭环数据)');
|
||
parts.push('');
|
||
parts.push('输出格式:');
|
||
parts.push('---');
|
||
parts.push('## 📡 SYSLOG 验收报告');
|
||
parts.push('### 广播编号:[编号]');
|
||
parts.push('### 验收结果:[通过/需补充]');
|
||
parts.push('### 模块验证:[已上传/未上传,列出详情]');
|
||
parts.push('### 工作总结:[摘要]');
|
||
parts.push('### 画像评估:[PGP 五维度评分]');
|
||
parts.push('### 调度判断:[RT-02 下一步]');
|
||
parts.push('### 反馈与建议:[内容]');
|
||
parts.push('---');
|
||
} else {
|
||
parts.push('');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('## 🎯 当前任务:开发者提问解答');
|
||
parts.push('═══════════════════════════════════════════');
|
||
parts.push('');
|
||
parts.push('你需要完成以下工作:');
|
||
parts.push('1. 理解开发者的问题');
|
||
parts.push('2. 结合广播上下文和开发者画像思考');
|
||
parts.push('3. 给出清晰、可操作的解答');
|
||
parts.push('4. 如果问题涉及代码,提供代码示例');
|
||
parts.push('');
|
||
parts.push('输出格式:');
|
||
parts.push('---');
|
||
parts.push('## 💡 问题解答');
|
||
parts.push('### 广播编号:[编号]');
|
||
parts.push('### 问题理解:[你对问题的理解]');
|
||
parts.push('### 解答:[详细解答]');
|
||
parts.push('### 建议:[后续建议]');
|
||
parts.push('---');
|
||
}
|
||
|
||
var prompt = parts.join('\n');
|
||
console.log('[Prompt] → System prompt 构建完成: ' + prompt.length + ' 字符');
|
||
return prompt;
|
||
}
|
||
|
||
// ══════════════════════════════════════════════════════════
|
||
// 主流程
|
||
// ══════════════════════════════════════════════════════════
|
||
|
||
async function main() {
|
||
console.log('═══════════════════════════════════════════');
|
||
console.log('🧠 铸渊 · 人格体唤醒管道');
|
||
console.log('═══════════════════════════════════════════');
|
||
console.log(' 广播编号: ' + BROADCAST_ID);
|
||
console.log(' 类型: ' + SUBMIT_TYPE);
|
||
console.log(' 提交者: ' + AUTHOR);
|
||
console.log(' 平台: ' + LLM_BASE_URL);
|
||
console.log(' 内容长度: ' + SUBMIT_CONTENT.length + ' 字符');
|
||
console.log('');
|
||
|
||
// 构建 prompts(动态注入 v4.0 协议)
|
||
const systemPrompt = await buildSystemPrompt(SUBMIT_TYPE, BROADCAST_ID, AUTHOR);
|
||
const userMessage = SUBMIT_CONTENT;
|
||
|
||
// 调用 LLM
|
||
console.log('🧠 正在唤醒人格体...');
|
||
const result = await callLLM(systemPrompt, userMessage);
|
||
console.log('');
|
||
console.log('✅ 人格体处理完成');
|
||
console.log(' 结果长度: ' + result.length + ' 字符');
|
||
|
||
// 输出结果到 GitHub Actions output
|
||
// 使用 GITHUB_OUTPUT 环境文件(支持多行)
|
||
const outputFile = process.env.GITHUB_OUTPUT;
|
||
if (outputFile) {
|
||
const delimiter = 'EOF_' + Date.now();
|
||
fs.appendFileSync(outputFile, 'result<<' + delimiter + '\n' + result + '\n' + delimiter + '\n');
|
||
}
|
||
|
||
// 输出广播结果到 /tmp/broadcast-result.json(供飞书推送步骤使用)
|
||
try {
|
||
const broadcastResult = {
|
||
title: '',
|
||
summary: '',
|
||
notion_url: '',
|
||
dev_id: AUTHOR,
|
||
raw_length: result.length,
|
||
timestamp: new Date().toISOString(),
|
||
};
|
||
|
||
// 尝试从人格体输出中提取广播标题(BC-xxx 格式)
|
||
const bcMatch = result.match(/BC-[\w]+-\d+-[\w]+/);
|
||
if (bcMatch) {
|
||
broadcastResult.title = bcMatch[0];
|
||
}
|
||
|
||
// 尝试提取摘要(取前100字符)
|
||
const summaryLines = result.split('\n').filter(function (l) { return l.trim().length > 0; });
|
||
if (summaryLines.length > 0) {
|
||
broadcastResult.summary = summaryLines[0].slice(0, 100);
|
||
}
|
||
|
||
fs.writeFileSync('/tmp/broadcast-result.json', JSON.stringify(broadcastResult, null, 2));
|
||
console.log(' 📄 broadcast-result.json 已写入 /tmp/');
|
||
} catch (e) {
|
||
console.log(' ⚠️ broadcast-result.json 写入失败: ' + e.message);
|
||
}
|
||
|
||
// 同时输出到 stdout 供调试
|
||
console.log('');
|
||
console.log('═══════════════════════════════════════════');
|
||
console.log('📋 人格体输出:');
|
||
console.log('═══════════════════════════════════════════');
|
||
console.log(result);
|
||
}
|
||
|
||
main().catch(function (err) {
|
||
console.error('❌ 人格体唤醒失败: ' + err.message);
|
||
// 即使 LLM 失败,也写一个 fallback 输出,让后续步骤可以继续
|
||
const outputFile = process.env.GITHUB_OUTPUT;
|
||
if (outputFile) {
|
||
const fallback = '(人格体唤醒失败: ' + err.message + ',请检查 LLM_API_KEY 和 LLM_BASE_URL 配置)';
|
||
const delimiter = 'EOF_' + Date.now();
|
||
fs.appendFileSync(outputFile, 'result<<' + delimiter + '\n' + fallback + '\n' + delimiter + '\n');
|
||
}
|
||
process.exit(1);
|
||
});
|