deepspider 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.claude/agents/check.md +122 -0
- package/.claude/agents/debug.md +106 -0
- package/.claude/agents/dispatch.md +214 -0
- package/.claude/agents/implement.md +96 -0
- package/.claude/agents/plan.md +396 -0
- package/.claude/agents/research.md +120 -0
- package/.claude/commands/evolve/merge.md +80 -0
- package/.claude/commands/trellis/before-backend-dev.md +13 -0
- package/.claude/commands/trellis/before-frontend-dev.md +13 -0
- package/.claude/commands/trellis/break-loop.md +107 -0
- package/.claude/commands/trellis/check-backend.md +13 -0
- package/.claude/commands/trellis/check-cross-layer.md +153 -0
- package/.claude/commands/trellis/check-frontend.md +13 -0
- package/.claude/commands/trellis/create-command.md +154 -0
- package/.claude/commands/trellis/finish-work.md +129 -0
- package/.claude/commands/trellis/integrate-skill.md +219 -0
- package/.claude/commands/trellis/onboard.md +358 -0
- package/.claude/commands/trellis/parallel.md +193 -0
- package/.claude/commands/trellis/record-session.md +62 -0
- package/.claude/commands/trellis/start.md +280 -0
- package/.claude/commands/trellis/update-spec.md +213 -0
- package/.claude/hooks/inject-subagent-context.py +758 -0
- package/.claude/hooks/ralph-loop.py +374 -0
- package/.claude/hooks/session-start.py +126 -0
- package/.claude/settings.json +41 -0
- package/.claude/skills/deepagents-guide/SKILL.md +428 -0
- package/.cursor/commands/trellis-before-backend-dev.md +13 -0
- package/.cursor/commands/trellis-before-frontend-dev.md +13 -0
- package/.cursor/commands/trellis-break-loop.md +107 -0
- package/.cursor/commands/trellis-check-backend.md +13 -0
- package/.cursor/commands/trellis-check-cross-layer.md +153 -0
- package/.cursor/commands/trellis-check-frontend.md +13 -0
- package/.cursor/commands/trellis-create-command.md +154 -0
- package/.cursor/commands/trellis-finish-work.md +129 -0
- package/.cursor/commands/trellis-integrate-skill.md +219 -0
- package/.cursor/commands/trellis-onboard.md +358 -0
- package/.cursor/commands/trellis-record-session.md +62 -0
- package/.cursor/commands/trellis-start.md +156 -0
- package/.cursor/commands/trellis-update-spec.md +213 -0
- package/.env.example +11 -0
- package/.husky/pre-commit +1 -0
- package/.mcp.json +8 -0
- package/.trellis/.template-hashes.json +65 -0
- package/.trellis/.version +1 -0
- package/.trellis/scripts/add-session.sh +384 -0
- package/.trellis/scripts/common/developer.sh +129 -0
- package/.trellis/scripts/common/git-context.sh +263 -0
- package/.trellis/scripts/common/paths.sh +208 -0
- package/.trellis/scripts/common/phase.sh +150 -0
- package/.trellis/scripts/common/registry.sh +247 -0
- package/.trellis/scripts/common/task-queue.sh +142 -0
- package/.trellis/scripts/common/task-utils.sh +151 -0
- package/.trellis/scripts/common/worktree.sh +128 -0
- package/.trellis/scripts/create-bootstrap.sh +299 -0
- package/.trellis/scripts/get-context.sh +7 -0
- package/.trellis/scripts/get-developer.sh +15 -0
- package/.trellis/scripts/init-developer.sh +34 -0
- package/.trellis/scripts/multi-agent/cleanup.sh +396 -0
- package/.trellis/scripts/multi-agent/create-pr.sh +241 -0
- package/.trellis/scripts/multi-agent/plan.sh +207 -0
- package/.trellis/scripts/multi-agent/start.sh +310 -0
- package/.trellis/scripts/multi-agent/status.sh +828 -0
- package/.trellis/scripts/task.sh +1118 -0
- package/.trellis/spec/backend/deepagents-guide.md +337 -0
- package/.trellis/spec/backend/directory-structure.md +126 -0
- package/.trellis/spec/backend/examples/skills/deepagents-guide/README.md +11 -0
- package/.trellis/spec/backend/examples/skills/deepagents-guide/agent.js.template +20 -0
- package/.trellis/spec/backend/examples/skills/deepagents-guide/skills-config.js.template +13 -0
- package/.trellis/spec/backend/examples/skills/deepagents-guide/subagent.js.template +19 -0
- package/.trellis/spec/backend/hook-guidelines.md +178 -0
- package/.trellis/spec/backend/index.md +36 -0
- package/.trellis/spec/backend/quality-guidelines.md +201 -0
- package/.trellis/spec/backend/state-management.md +76 -0
- package/.trellis/spec/backend/tool-guidelines.md +144 -0
- package/.trellis/spec/backend/type-safety.md +71 -0
- package/.trellis/spec/guides/code-reuse-thinking-guide.md +92 -0
- package/.trellis/spec/guides/cross-layer-thinking-guide.md +94 -0
- package/.trellis/spec/guides/index.md +79 -0
- package/.trellis/tasks/archive/02-02-evolving-skills/prd.md +61 -0
- package/.trellis/tasks/archive/02-02-evolving-skills/task.json +29 -0
- package/.trellis/tasks/archive/2026-02/00-bootstrap-guidelines/prd.md +86 -0
- package/.trellis/tasks/archive/2026-02/00-bootstrap-guidelines/task.json +27 -0
- package/.trellis/tasks/archive/2026-02/02-02-skills-system/check.jsonl +3 -0
- package/.trellis/tasks/archive/2026-02/02-02-skills-system/debug.jsonl +2 -0
- package/.trellis/tasks/archive/2026-02/02-02-skills-system/implement.jsonl +5 -0
- package/.trellis/tasks/archive/2026-02/02-02-skills-system/prd.md +33 -0
- package/.trellis/tasks/archive/2026-02/02-02-skills-system/task.json +41 -0
- package/.trellis/workflow.md +407 -0
- package/.trellis/workspace/index.md +123 -0
- package/.trellis/workspace/pony/index.md +40 -0
- package/.trellis/workspace/pony/journal-1.md +7 -0
- package/.trellis/worktree.yaml +47 -0
- package/AGENTS.md +18 -0
- package/CLAUDE.md +292 -0
- package/README.md +134 -0
- package/agents/deepspider.md +142 -0
- package/docs/DEBUG.md +42 -0
- package/docs/GUIDE.md +334 -0
- package/docs/PROMPT.md +60 -0
- package/docs/USAGE.md +226 -0
- package/eslint.config.js +51 -0
- package/package.json +78 -0
- package/requirements-crypto.txt +14 -0
- package/src/agent/index.js +97 -0
- package/src/agent/logger.js +164 -0
- package/src/agent/middleware/filterTools.js +64 -0
- package/src/agent/middleware/report.js +79 -0
- package/src/agent/prompts/system.js +315 -0
- package/src/agent/run.js +575 -0
- package/src/agent/skills/anti-detect/SKILL.md +28 -0
- package/src/agent/skills/anti-detect/evolved.md +12 -0
- package/src/agent/skills/captcha/SKILL.md +37 -0
- package/src/agent/skills/captcha/evolved.md +12 -0
- package/src/agent/skills/config.js +30 -0
- package/src/agent/skills/crawler/SKILL.md +9 -0
- package/src/agent/skills/crawler/evolved.md +16 -0
- package/src/agent/skills/dynamic-analysis/SKILL.md +91 -0
- package/src/agent/skills/dynamic-analysis/evolved.md +12 -0
- package/src/agent/skills/env/SKILL.md +72 -0
- package/src/agent/skills/env/evolved.md +12 -0
- package/src/agent/skills/evolve.js +79 -0
- package/src/agent/skills/general/SKILL.md +12 -0
- package/src/agent/skills/general/evolved.md +12 -0
- package/src/agent/skills/js2python/SKILL.md +30 -0
- package/src/agent/skills/js2python/evolved.md +13 -0
- package/src/agent/skills/report/SKILL.md +21 -0
- package/src/agent/skills/report/evolved.md +12 -0
- package/src/agent/skills/sandbox/SKILL.md +22 -0
- package/src/agent/skills/sandbox/evolved.md +16 -0
- package/src/agent/skills/static-analysis/SKILL.md +93 -0
- package/src/agent/skills/static-analysis/evolved.md +12 -0
- package/src/agent/skills/xpath/SKILL.md +119 -0
- package/src/agent/subagents/anti-detect.js +45 -0
- package/src/agent/subagents/captcha.js +51 -0
- package/src/agent/subagents/crawler.js +138 -0
- package/src/agent/subagents/dynamic.js +64 -0
- package/src/agent/subagents/env-agent.js +82 -0
- package/src/agent/subagents/index.js +37 -0
- package/src/agent/subagents/js2python.js +72 -0
- package/src/agent/subagents/sandbox.js +55 -0
- package/src/agent/subagents/static.js +66 -0
- package/src/agent/tools/analysis.js +135 -0
- package/src/agent/tools/analyzer.js +85 -0
- package/src/agent/tools/anti-detect.js +89 -0
- package/src/agent/tools/antidebug.js +64 -0
- package/src/agent/tools/async.js +43 -0
- package/src/agent/tools/browser.js +324 -0
- package/src/agent/tools/captcha.js +223 -0
- package/src/agent/tools/capture.js +179 -0
- package/src/agent/tools/correlate.js +303 -0
- package/src/agent/tools/crawler.js +116 -0
- package/src/agent/tools/cryptohook.js +80 -0
- package/src/agent/tools/debug.js +246 -0
- package/src/agent/tools/deobfuscator.js +90 -0
- package/src/agent/tools/env.js +83 -0
- package/src/agent/tools/envdump.js +92 -0
- package/src/agent/tools/evolve.js +164 -0
- package/src/agent/tools/extract.js +114 -0
- package/src/agent/tools/extractor.js +54 -0
- package/src/agent/tools/file.js +224 -0
- package/src/agent/tools/hook.js +84 -0
- package/src/agent/tools/hookManager.js +178 -0
- package/src/agent/tools/index.js +137 -0
- package/src/agent/tools/nodejs.js +101 -0
- package/src/agent/tools/patch.js +46 -0
- package/src/agent/tools/preprocess.js +71 -0
- package/src/agent/tools/profile.js +122 -0
- package/src/agent/tools/python.js +627 -0
- package/src/agent/tools/report.js +124 -0
- package/src/agent/tools/runtime.js +132 -0
- package/src/agent/tools/sandbox.js +79 -0
- package/src/agent/tools/store.js +73 -0
- package/src/agent/tools/trace.js +74 -0
- package/src/agent/tools/tracing.js +201 -0
- package/src/agent/tools/utils.js +51 -0
- package/src/agent/tools/verify.js +184 -0
- package/src/agent/tools/webcrack.js +109 -0
- package/src/analyzer/ASTAnalyzer.js +387 -0
- package/src/analyzer/CallStackAnalyzer.js +379 -0
- package/src/analyzer/Deobfuscator.js +289 -0
- package/src/analyzer/EncryptionAnalyzer.js +99 -0
- package/src/analyzer/index.js +22 -0
- package/src/browser/EnvBridge.js +186 -0
- package/src/browser/cdp.js +168 -0
- package/src/browser/client.js +197 -0
- package/src/browser/collector.js +444 -0
- package/src/browser/collectors/RequestCryptoLinker.js +109 -0
- package/src/browser/collectors/ResponseSearcher.js +107 -0
- package/src/browser/collectors/ScriptCollector.js +158 -0
- package/src/browser/collectors/index.js +26 -0
- package/src/browser/defaultHooks.js +932 -0
- package/src/browser/hooks/crypto.js +55 -0
- package/src/browser/hooks/index.js +64 -0
- package/src/browser/hooks/native.js +9 -0
- package/src/browser/hooks/network.js +33 -0
- package/src/browser/index.js +42 -0
- package/src/browser/interceptors/NetworkInterceptor.js +116 -0
- package/src/browser/interceptors/ScriptInterceptor.js +76 -0
- package/src/browser/interceptors/index.js +6 -0
- package/src/browser/ui/analysisPanel.js +1782 -0
- package/src/browser/ui/confirmDialog.js +158 -0
- package/src/browser/ui/panel.html +152 -0
- package/src/browser/ui/selector.js +170 -0
- package/src/config/index.js +5 -0
- package/src/config/paths.js +71 -0
- package/src/config/patterns/crypto.js +36 -0
- package/src/config/profiles/chrome.json +71 -0
- package/src/config/profiles/firefox.json +44 -0
- package/src/config/profiles/safari.json +38 -0
- package/src/core/EnvMonitor.js +200 -0
- package/src/core/PatchGenerator.js +278 -0
- package/src/core/Sandbox.js +181 -0
- package/src/env/AntiAntiDebug.js +111 -0
- package/src/env/AsyncHook.js +68 -0
- package/src/env/BrowserAPIList.js +265 -0
- package/src/env/CookieHook.js +48 -0
- package/src/env/CryptoHook.js +205 -0
- package/src/env/EnvCodeGenerator.js +157 -0
- package/src/env/EnvDumper.js +356 -0
- package/src/env/EnvExtractor.js +220 -0
- package/src/env/HookBase.js +618 -0
- package/src/env/NetworkHook.js +159 -0
- package/src/env/modules/bom/history.js +29 -0
- package/src/env/modules/bom/location.js +26 -0
- package/src/env/modules/bom/navigator.js +70 -0
- package/src/env/modules/bom/screen.js +26 -0
- package/src/env/modules/bom/storage.js +23 -0
- package/src/env/modules/dom/document.js +110 -0
- package/src/env/modules/dom/event.js +51 -0
- package/src/env/modules/index.js +34 -0
- package/src/env/modules/webapi/fetch.js +46 -0
- package/src/env/modules/webapi/url.js +47 -0
- package/src/env/modules/webapi/xhr.js +48 -0
- package/src/index.js +27 -0
- package/src/mcp/server.js +89 -0
- package/src/store/DataStore.js +708 -0
- package/src/store/Store.js +158 -0
- package/src/store/Validator.js +24 -0
- package/test/analyze.test.js +90 -0
- package/test/envdump.test.js +74 -0
- package/test/flow.test.js +90 -0
- package/test/hooks.test.js +138 -0
- package/test/plugin.test.js +35 -0
- package/test/refactor-full.test.js +30 -0
- package/test/refactor.test.js +21 -0
- package/test/samples/obfuscated.js +61 -0
- package/test/samples/original.js +66 -0
- package/test/samples/v10_eval_chain.js +52 -0
- package/test/samples/v11_bytecode_vm.js +81 -0
- package/test/samples/v12_polymorphic.js +69 -0
- package/test/samples/v1_ob_basic.js +98 -0
- package/test/samples/v2_ob_advanced.js +99 -0
- package/test/samples/v3_jjencode.js +77 -0
- package/test/samples/v4_aaencode.js +73 -0
- package/test/samples/v5_control_flow.js +86 -0
- package/test/samples/v6_string_encryption.js +71 -0
- package/test/samples/v7_jsvmp.js +83 -0
- package/test/samples/v8_anti_debug.js +79 -0
- package/test/samples/v9_proxy_trap.js +49 -0
- package/test/samples.test.js +96 -0
- package/test/webcrack.test.js +55 -0
|
@@ -0,0 +1,99 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* DeepSpider - 加密分析器
|
|
3
|
+
* 算法识别、参数追踪、密钥提取
|
|
4
|
+
*/
|
|
5
|
+
|
|
6
|
+
import { ASTAnalyzer } from './ASTAnalyzer.js';
|
|
7
|
+
import { cryptoPatterns } from '../config/patterns/crypto.js';
|
|
8
|
+
|
|
9
|
+
export class EncryptionAnalyzer {
|
|
10
|
+
constructor() {
|
|
11
|
+
this.astAnalyzer = new ASTAnalyzer();
|
|
12
|
+
this.cryptoPatterns = [
|
|
13
|
+
{ name: 'MD5', pattern: /md5|MD5/, type: 'hash' },
|
|
14
|
+
{ name: 'SHA1', pattern: /sha1|SHA1/, type: 'hash' },
|
|
15
|
+
{ name: 'SHA256', pattern: /sha256|SHA256/, type: 'hash' },
|
|
16
|
+
{ name: 'AES', pattern: /\baes\b|AES|CryptoJS\.AES/, type: 'symmetric' },
|
|
17
|
+
{ name: 'DES', pattern: /\bdes\b|DES|CryptoJS\.DES/, type: 'symmetric' },
|
|
18
|
+
{ name: 'Base64', pattern: /btoa|atob|base64/i, type: 'encoding' },
|
|
19
|
+
{ name: 'RSA', pattern: /\brsa\b|RSA|JSEncrypt/, type: 'asymmetric' },
|
|
20
|
+
{ name: 'HMAC', pattern: /hmac|HMAC/, type: 'mac' }
|
|
21
|
+
];
|
|
22
|
+
}
|
|
23
|
+
|
|
24
|
+
analyze(code) {
|
|
25
|
+
const functions = this.astAnalyzer.extractFunctions(code);
|
|
26
|
+
const calls = this.astAnalyzer.extractCalls(code);
|
|
27
|
+
const detected = [];
|
|
28
|
+
|
|
29
|
+
// 检测加密模式
|
|
30
|
+
for (const p of this.cryptoPatterns) {
|
|
31
|
+
if (p.pattern.test(code)) {
|
|
32
|
+
detected.push({ name: p.name, type: p.type });
|
|
33
|
+
}
|
|
34
|
+
}
|
|
35
|
+
|
|
36
|
+
// 查找可疑的加密函数
|
|
37
|
+
const suspiciousFuncs = functions.filter(f =>
|
|
38
|
+
/encrypt|decrypt|sign|hash|encode|decode|cipher/i.test(f.name)
|
|
39
|
+
);
|
|
40
|
+
|
|
41
|
+
// 查找密钥相关变量
|
|
42
|
+
const keys = this._findKeys(code);
|
|
43
|
+
|
|
44
|
+
return {
|
|
45
|
+
detectedAlgorithms: detected,
|
|
46
|
+
suspiciousFunctions: suspiciousFuncs,
|
|
47
|
+
possibleKeys: keys,
|
|
48
|
+
totalFunctions: functions.length,
|
|
49
|
+
totalCalls: calls.length
|
|
50
|
+
};
|
|
51
|
+
}
|
|
52
|
+
|
|
53
|
+
traceParam(code, paramName) {
|
|
54
|
+
const ast = this.astAnalyzer.parse(code);
|
|
55
|
+
const traces = [];
|
|
56
|
+
|
|
57
|
+
// 简化实现:查找参数使用位置
|
|
58
|
+
const regex = new RegExp(`\\b${paramName}\\b`, 'g');
|
|
59
|
+
let match;
|
|
60
|
+
while ((match = regex.exec(code)) !== null) {
|
|
61
|
+
traces.push({ position: match.index });
|
|
62
|
+
}
|
|
63
|
+
|
|
64
|
+
return { param: paramName, traces };
|
|
65
|
+
}
|
|
66
|
+
|
|
67
|
+
_findKeys(code) {
|
|
68
|
+
const keys = [];
|
|
69
|
+
const keyPattern = /(?:key|secret|password|iv|salt)\s*[=:]\s*['"]([^'"]+)['"]/gi;
|
|
70
|
+
let match;
|
|
71
|
+
while ((match = keyPattern.exec(code)) !== null) {
|
|
72
|
+
keys.push({
|
|
73
|
+
value: match[1],
|
|
74
|
+
position: match.index
|
|
75
|
+
});
|
|
76
|
+
}
|
|
77
|
+
return keys;
|
|
78
|
+
}
|
|
79
|
+
|
|
80
|
+
// 使用模式库进行深度检测
|
|
81
|
+
detectWithPatterns(code) {
|
|
82
|
+
const detected = [];
|
|
83
|
+
for (const [key, pattern] of Object.entries(cryptoPatterns)) {
|
|
84
|
+
for (const sig of pattern.signatures) {
|
|
85
|
+
if (sig.test(code)) {
|
|
86
|
+
detected.push({
|
|
87
|
+
algorithm: pattern.name,
|
|
88
|
+
type: pattern.type,
|
|
89
|
+
confidence: 0.9
|
|
90
|
+
});
|
|
91
|
+
break;
|
|
92
|
+
}
|
|
93
|
+
}
|
|
94
|
+
}
|
|
95
|
+
return detected;
|
|
96
|
+
}
|
|
97
|
+
}
|
|
98
|
+
|
|
99
|
+
export default EncryptionAnalyzer;
|
|
@@ -0,0 +1,22 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* DeepSpider - 分析器注册表
|
|
3
|
+
*/
|
|
4
|
+
|
|
5
|
+
import { ASTAnalyzer } from './ASTAnalyzer.js';
|
|
6
|
+
import { CallStackAnalyzer } from './CallStackAnalyzer.js';
|
|
7
|
+
import { EncryptionAnalyzer } from './EncryptionAnalyzer.js';
|
|
8
|
+
import { Deobfuscator } from './Deobfuscator.js';
|
|
9
|
+
|
|
10
|
+
export const analyzers = {
|
|
11
|
+
ast: ASTAnalyzer,
|
|
12
|
+
callstack: CallStackAnalyzer,
|
|
13
|
+
encryption: EncryptionAnalyzer,
|
|
14
|
+
deobfuscator: Deobfuscator
|
|
15
|
+
};
|
|
16
|
+
|
|
17
|
+
export function getAnalyzer(name) {
|
|
18
|
+
const Analyzer = analyzers[name];
|
|
19
|
+
return Analyzer ? new Analyzer() : null;
|
|
20
|
+
}
|
|
21
|
+
|
|
22
|
+
export default analyzers;
|
|
@@ -0,0 +1,186 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* DeepSpider - 环境补全桥接器
|
|
3
|
+
* 连接 EnvDumper(发现缺失)与 EnvCollector(采集真实数据)
|
|
4
|
+
*/
|
|
5
|
+
|
|
6
|
+
import { EnvCollector } from './collector.js';
|
|
7
|
+
import { PatchGenerator } from '../core/PatchGenerator.js';
|
|
8
|
+
|
|
9
|
+
export class EnvBridge {
|
|
10
|
+
constructor(page) {
|
|
11
|
+
this.collector = new EnvCollector(page);
|
|
12
|
+
this.patchGenerator = new PatchGenerator();
|
|
13
|
+
this.collectedData = new Map();
|
|
14
|
+
}
|
|
15
|
+
|
|
16
|
+
/**
|
|
17
|
+
* 根据缺失列表自动采集并生成补丁
|
|
18
|
+
* @param {string[]} missingPaths - EnvDumper 发现的缺失属性路径
|
|
19
|
+
*/
|
|
20
|
+
async autoFix(missingPaths) {
|
|
21
|
+
const results = {
|
|
22
|
+
collected: [],
|
|
23
|
+
patched: [],
|
|
24
|
+
failed: []
|
|
25
|
+
};
|
|
26
|
+
|
|
27
|
+
for (const path of missingPaths) {
|
|
28
|
+
try {
|
|
29
|
+
// 1. 从真实浏览器采集
|
|
30
|
+
const collected = await this.collector.collect(path, { depth: 2 });
|
|
31
|
+
|
|
32
|
+
if (!collected.success) {
|
|
33
|
+
results.failed.push({ path, reason: 'collect_failed', error: collected.error });
|
|
34
|
+
continue;
|
|
35
|
+
}
|
|
36
|
+
|
|
37
|
+
results.collected.push(path);
|
|
38
|
+
this.collectedData.set(path, collected);
|
|
39
|
+
|
|
40
|
+
// 2. 生成补丁代码
|
|
41
|
+
const patch = this._generatePatch(path, collected);
|
|
42
|
+
if (patch) {
|
|
43
|
+
results.patched.push({ path, code: patch });
|
|
44
|
+
}
|
|
45
|
+
|
|
46
|
+
} catch (e) {
|
|
47
|
+
results.failed.push({ path, reason: 'error', error: e.message });
|
|
48
|
+
}
|
|
49
|
+
}
|
|
50
|
+
|
|
51
|
+
return results;
|
|
52
|
+
}
|
|
53
|
+
|
|
54
|
+
/**
|
|
55
|
+
* 根据采集数据生成补丁代码
|
|
56
|
+
*/
|
|
57
|
+
_generatePatch(path, collected) {
|
|
58
|
+
const { data, descriptor } = collected;
|
|
59
|
+
|
|
60
|
+
if (!data) return null;
|
|
61
|
+
|
|
62
|
+
const parts = path.split('.');
|
|
63
|
+
const propName = parts.pop();
|
|
64
|
+
const parentPath = parts.join('.') || 'window';
|
|
65
|
+
|
|
66
|
+
// 根据数据类型生成不同的补丁
|
|
67
|
+
switch (data.type) {
|
|
68
|
+
case 'string':
|
|
69
|
+
return `${parentPath}.${propName} = "${data.value}";`;
|
|
70
|
+
|
|
71
|
+
case 'number':
|
|
72
|
+
return `${parentPath}.${propName} = ${data.value};`;
|
|
73
|
+
|
|
74
|
+
case 'boolean':
|
|
75
|
+
return `${parentPath}.${propName} = ${data.value};`;
|
|
76
|
+
|
|
77
|
+
case 'array':
|
|
78
|
+
return `${parentPath}.${propName} = ${JSON.stringify(
|
|
79
|
+
data.value.map(v => v.value)
|
|
80
|
+
)};`;
|
|
81
|
+
|
|
82
|
+
case 'object':
|
|
83
|
+
if (descriptor?.hasGetter) {
|
|
84
|
+
// 有 getter 的属性用 defineProperty
|
|
85
|
+
return this._generateDefineProperty(parentPath, propName, data);
|
|
86
|
+
}
|
|
87
|
+
return this._generateObjectPatch(parentPath, propName, data);
|
|
88
|
+
|
|
89
|
+
case 'function':
|
|
90
|
+
return `${parentPath}.${propName} = function ${data.name || ''}() {};`;
|
|
91
|
+
|
|
92
|
+
default:
|
|
93
|
+
return `${parentPath}.${propName} = undefined;`;
|
|
94
|
+
}
|
|
95
|
+
}
|
|
96
|
+
|
|
97
|
+
_generateDefineProperty(parentPath, propName, data) {
|
|
98
|
+
const value = this._serializeValue(data);
|
|
99
|
+
return `Object.defineProperty(${parentPath}, "${propName}", {
|
|
100
|
+
get: function() { return ${value}; },
|
|
101
|
+
configurable: true
|
|
102
|
+
});`;
|
|
103
|
+
}
|
|
104
|
+
|
|
105
|
+
_generateObjectPatch(parentPath, propName, data) {
|
|
106
|
+
if (!data.properties) {
|
|
107
|
+
return `${parentPath}.${propName} = {};`;
|
|
108
|
+
}
|
|
109
|
+
|
|
110
|
+
const props = [];
|
|
111
|
+
for (const [key, val] of Object.entries(data.properties)) {
|
|
112
|
+
if (val.type === 'function') {
|
|
113
|
+
props.push(`${key}: function() {}`);
|
|
114
|
+
} else if (val.type === 'object') {
|
|
115
|
+
props.push(`${key}: {}`);
|
|
116
|
+
} else {
|
|
117
|
+
props.push(`${key}: ${this._serializeValue(val)}`);
|
|
118
|
+
}
|
|
119
|
+
}
|
|
120
|
+
|
|
121
|
+
return `${parentPath}.${propName} = { ${props.join(', ')} };`;
|
|
122
|
+
}
|
|
123
|
+
|
|
124
|
+
_serializeValue(data) {
|
|
125
|
+
switch (data.type) {
|
|
126
|
+
case 'string': return `"${data.value}"`;
|
|
127
|
+
case 'number': return data.value;
|
|
128
|
+
case 'boolean': return data.value;
|
|
129
|
+
case 'null': return 'null';
|
|
130
|
+
case 'undefined': return 'undefined';
|
|
131
|
+
case 'array': return JSON.stringify(data.value?.map(v => v.value) || []);
|
|
132
|
+
default: return '{}';
|
|
133
|
+
}
|
|
134
|
+
}
|
|
135
|
+
|
|
136
|
+
/**
|
|
137
|
+
* 生成合并后的完整补丁代码
|
|
138
|
+
*/
|
|
139
|
+
generateMergedPatch(patches) {
|
|
140
|
+
const grouped = new Map();
|
|
141
|
+
|
|
142
|
+
for (const { path, code } of patches) {
|
|
143
|
+
const root = path.split('.')[0];
|
|
144
|
+
if (!grouped.has(root)) {
|
|
145
|
+
grouped.set(root, []);
|
|
146
|
+
}
|
|
147
|
+
grouped.get(root).push(code);
|
|
148
|
+
}
|
|
149
|
+
|
|
150
|
+
const lines = ['// DeepSpider Auto-Generated Patch'];
|
|
151
|
+
for (const [root, codes] of grouped) {
|
|
152
|
+
lines.push(`\n// === ${root} ===`);
|
|
153
|
+
lines.push(...codes);
|
|
154
|
+
}
|
|
155
|
+
|
|
156
|
+
return lines.join('\n');
|
|
157
|
+
}
|
|
158
|
+
|
|
159
|
+
/**
|
|
160
|
+
* 完整的补环境流程
|
|
161
|
+
*/
|
|
162
|
+
async runFullPipeline(missingPaths) {
|
|
163
|
+
console.log(`[DeepSpider:bridge] 开始补环境,共 ${missingPaths.length} 个缺失属性`);
|
|
164
|
+
|
|
165
|
+
// 1. 采集并生成补丁
|
|
166
|
+
const result = await this.autoFix(missingPaths);
|
|
167
|
+
|
|
168
|
+
// 2. 合并补丁
|
|
169
|
+
const mergedCode = this.generateMergedPatch(result.patched);
|
|
170
|
+
|
|
171
|
+
// 3. 返回结果
|
|
172
|
+
return {
|
|
173
|
+
success: result.failed.length === 0,
|
|
174
|
+
stats: {
|
|
175
|
+
total: missingPaths.length,
|
|
176
|
+
collected: result.collected.length,
|
|
177
|
+
patched: result.patched.length,
|
|
178
|
+
failed: result.failed.length
|
|
179
|
+
},
|
|
180
|
+
patchCode: mergedCode,
|
|
181
|
+
failed: result.failed
|
|
182
|
+
};
|
|
183
|
+
}
|
|
184
|
+
}
|
|
185
|
+
|
|
186
|
+
export default EnvBridge;
|
|
@@ -0,0 +1,168 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* DeepSpider - CDP 会话管理
|
|
3
|
+
* 封装 Chrome DevTools Protocol 操作
|
|
4
|
+
*/
|
|
5
|
+
|
|
6
|
+
export class CDPSession {
|
|
7
|
+
constructor(client) {
|
|
8
|
+
this.client = client;
|
|
9
|
+
this.breakpoints = new Map();
|
|
10
|
+
this.scriptSources = new Map();
|
|
11
|
+
this.pausedHandler = null;
|
|
12
|
+
}
|
|
13
|
+
|
|
14
|
+
/**
|
|
15
|
+
* 从 BrowserClient 创建 CDP 会话
|
|
16
|
+
*/
|
|
17
|
+
static async fromBrowser(browserClient) {
|
|
18
|
+
const client = await browserClient.getCDPSession();
|
|
19
|
+
const session = new CDPSession(client);
|
|
20
|
+
await session.enable();
|
|
21
|
+
return session;
|
|
22
|
+
}
|
|
23
|
+
|
|
24
|
+
/**
|
|
25
|
+
* 启用必要的 CDP 域
|
|
26
|
+
*/
|
|
27
|
+
async enable() {
|
|
28
|
+
await Promise.all([
|
|
29
|
+
this.client.send('Debugger.enable'),
|
|
30
|
+
this.client.send('Runtime.enable'),
|
|
31
|
+
this.client.send('Page.enable'),
|
|
32
|
+
this.client.send('Network.enable'),
|
|
33
|
+
]);
|
|
34
|
+
|
|
35
|
+
// 设置异步调用栈深度
|
|
36
|
+
await this.client.send('Debugger.setAsyncCallStackDepth', { maxDepth: 32 });
|
|
37
|
+
|
|
38
|
+
// 监听脚本解析事件
|
|
39
|
+
this.client.on('Debugger.scriptParsed', (event) => {
|
|
40
|
+
if (event.url) {
|
|
41
|
+
this.scriptSources.set(event.scriptId, event.url);
|
|
42
|
+
}
|
|
43
|
+
});
|
|
44
|
+
}
|
|
45
|
+
|
|
46
|
+
/**
|
|
47
|
+
* 设置断点
|
|
48
|
+
*/
|
|
49
|
+
async setBreakpoint(url, lineNumber, columnNumber = 0) {
|
|
50
|
+
const result = await this.client.send('Debugger.setBreakpointByUrl', {
|
|
51
|
+
url,
|
|
52
|
+
lineNumber,
|
|
53
|
+
columnNumber,
|
|
54
|
+
});
|
|
55
|
+
if (result.breakpointId) {
|
|
56
|
+
this.breakpoints.set(result.breakpointId, { url, lineNumber, columnNumber });
|
|
57
|
+
}
|
|
58
|
+
return result;
|
|
59
|
+
}
|
|
60
|
+
|
|
61
|
+
/**
|
|
62
|
+
* 移除断点
|
|
63
|
+
*/
|
|
64
|
+
async removeBreakpoint(breakpointId) {
|
|
65
|
+
await this.client.send('Debugger.removeBreakpoint', { breakpointId });
|
|
66
|
+
this.breakpoints.delete(breakpointId);
|
|
67
|
+
}
|
|
68
|
+
|
|
69
|
+
/**
|
|
70
|
+
* 设置 XHR 断点
|
|
71
|
+
*/
|
|
72
|
+
async setXHRBreakpoint(urlPattern = '') {
|
|
73
|
+
await this.client.send('DOMDebugger.setXHRBreakpoint', { url: urlPattern });
|
|
74
|
+
}
|
|
75
|
+
|
|
76
|
+
/**
|
|
77
|
+
* 恢复执行
|
|
78
|
+
*/
|
|
79
|
+
async resume() {
|
|
80
|
+
await this.client.send('Debugger.resume');
|
|
81
|
+
}
|
|
82
|
+
|
|
83
|
+
/**
|
|
84
|
+
* 单步执行
|
|
85
|
+
*/
|
|
86
|
+
async stepOver() {
|
|
87
|
+
await this.client.send('Debugger.stepOver');
|
|
88
|
+
}
|
|
89
|
+
|
|
90
|
+
async stepInto() {
|
|
91
|
+
await this.client.send('Debugger.stepInto');
|
|
92
|
+
}
|
|
93
|
+
|
|
94
|
+
async stepOut() {
|
|
95
|
+
await this.client.send('Debugger.stepOut');
|
|
96
|
+
}
|
|
97
|
+
|
|
98
|
+
/**
|
|
99
|
+
* 获取脚本源码
|
|
100
|
+
*/
|
|
101
|
+
async getScriptSource(scriptId) {
|
|
102
|
+
const result = await this.client.send('Debugger.getScriptSource', { scriptId });
|
|
103
|
+
return result.scriptSource;
|
|
104
|
+
}
|
|
105
|
+
|
|
106
|
+
/**
|
|
107
|
+
* 获取作用域变量
|
|
108
|
+
*/
|
|
109
|
+
async getScopeVariables(callFrameId) {
|
|
110
|
+
const result = await this.client.send('Debugger.evaluateOnCallFrame', {
|
|
111
|
+
callFrameId,
|
|
112
|
+
expression: 'JSON.stringify(Object.keys(this))',
|
|
113
|
+
returnByValue: true,
|
|
114
|
+
});
|
|
115
|
+
return result;
|
|
116
|
+
}
|
|
117
|
+
|
|
118
|
+
/**
|
|
119
|
+
* 在调用帧上执行表达式
|
|
120
|
+
*/
|
|
121
|
+
async evaluate(callFrameId, expression) {
|
|
122
|
+
const result = await this.client.send('Debugger.evaluateOnCallFrame', {
|
|
123
|
+
callFrameId,
|
|
124
|
+
expression,
|
|
125
|
+
returnByValue: true,
|
|
126
|
+
});
|
|
127
|
+
return result.result?.value;
|
|
128
|
+
}
|
|
129
|
+
|
|
130
|
+
/**
|
|
131
|
+
* 监听断点暂停事件
|
|
132
|
+
*/
|
|
133
|
+
onPaused(handler) {
|
|
134
|
+
this.pausedHandler = handler;
|
|
135
|
+
this.client.on('Debugger.paused', handler);
|
|
136
|
+
}
|
|
137
|
+
|
|
138
|
+
/**
|
|
139
|
+
* 移除暂停监听
|
|
140
|
+
*/
|
|
141
|
+
offPaused() {
|
|
142
|
+
if (this.pausedHandler) {
|
|
143
|
+
this.client.off('Debugger.paused', this.pausedHandler);
|
|
144
|
+
this.pausedHandler = null;
|
|
145
|
+
}
|
|
146
|
+
}
|
|
147
|
+
|
|
148
|
+
/**
|
|
149
|
+
* 发送 CDP 命令(代理到 client)
|
|
150
|
+
*/
|
|
151
|
+
send(method, params = {}) {
|
|
152
|
+
return this.client.send(method, params);
|
|
153
|
+
}
|
|
154
|
+
|
|
155
|
+
/**
|
|
156
|
+
* 监听 CDP 事件(代理到 client)
|
|
157
|
+
*/
|
|
158
|
+
on(event, handler) {
|
|
159
|
+
return this.client.on(event, handler);
|
|
160
|
+
}
|
|
161
|
+
|
|
162
|
+
/**
|
|
163
|
+
* 移除 CDP 事件监听(代理到 client)
|
|
164
|
+
*/
|
|
165
|
+
off(event, handler) {
|
|
166
|
+
return this.client.off(event, handler);
|
|
167
|
+
}
|
|
168
|
+
}
|
|
@@ -0,0 +1,197 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* DeepSpider - Patchright 浏览器客户端
|
|
3
|
+
* 使用反检测版 Playwright
|
|
4
|
+
*/
|
|
5
|
+
|
|
6
|
+
import { chromium } from 'patchright';
|
|
7
|
+
import { getDefaultHookScript } from './defaultHooks.js';
|
|
8
|
+
import { NetworkInterceptor } from './interceptors/NetworkInterceptor.js';
|
|
9
|
+
import { ScriptInterceptor } from './interceptors/ScriptInterceptor.js';
|
|
10
|
+
import { getDataStore } from '../store/DataStore.js';
|
|
11
|
+
|
|
12
|
+
export class BrowserClient {
|
|
13
|
+
constructor() {
|
|
14
|
+
this.browser = null;
|
|
15
|
+
this.context = null;
|
|
16
|
+
this.page = null;
|
|
17
|
+
this.pages = [];
|
|
18
|
+
this.cdpSession = null;
|
|
19
|
+
this.networkInterceptor = null;
|
|
20
|
+
this.scriptInterceptor = null;
|
|
21
|
+
this.hookScript = null;
|
|
22
|
+
this.onMessage = null;
|
|
23
|
+
}
|
|
24
|
+
|
|
25
|
+
/**
|
|
26
|
+
* 启动浏览器
|
|
27
|
+
*/
|
|
28
|
+
async launch(options = {}) {
|
|
29
|
+
// 启动新会话
|
|
30
|
+
const dataStore = getDataStore();
|
|
31
|
+
dataStore.startSession();
|
|
32
|
+
|
|
33
|
+
const {
|
|
34
|
+
headless = false,
|
|
35
|
+
executablePath = null,
|
|
36
|
+
args = [],
|
|
37
|
+
} = options;
|
|
38
|
+
|
|
39
|
+
const launchOptions = {
|
|
40
|
+
headless,
|
|
41
|
+
args: [
|
|
42
|
+
'--disable-blink-features=AutomationControlled',
|
|
43
|
+
'--disable-web-security',
|
|
44
|
+
'--ignore-certificate-errors',
|
|
45
|
+
...args,
|
|
46
|
+
],
|
|
47
|
+
};
|
|
48
|
+
|
|
49
|
+
if (executablePath) {
|
|
50
|
+
launchOptions.executablePath = executablePath;
|
|
51
|
+
}
|
|
52
|
+
|
|
53
|
+
this.browser = await chromium.launch(launchOptions);
|
|
54
|
+
this.context = await this.browser.newContext({
|
|
55
|
+
ignoreHTTPSErrors: true,
|
|
56
|
+
});
|
|
57
|
+
|
|
58
|
+
// 保存 hook 脚本
|
|
59
|
+
this.hookScript = getDefaultHookScript();
|
|
60
|
+
|
|
61
|
+
// 使用 addInitScript 在 context 级别注入
|
|
62
|
+
await this.context.addInitScript(this.hookScript);
|
|
63
|
+
|
|
64
|
+
this.page = await this.context.newPage();
|
|
65
|
+
|
|
66
|
+
// 监听新页面创建(弹窗、新标签页)
|
|
67
|
+
this.context.on('page', async (newPage) => {
|
|
68
|
+
console.log('[BrowserClient] 检测到新页面');
|
|
69
|
+
this.pages.push(newPage);
|
|
70
|
+
this.page = newPage; // 切换到新页面
|
|
71
|
+
await this.setupPage(newPage);
|
|
72
|
+
|
|
73
|
+
// 监听页面关闭
|
|
74
|
+
newPage.on('close', () => {
|
|
75
|
+
this.pages = this.pages.filter(p => p !== newPage);
|
|
76
|
+
if (this.page === newPage && this.pages.length > 0) {
|
|
77
|
+
this.page = this.pages[this.pages.length - 1];
|
|
78
|
+
}
|
|
79
|
+
});
|
|
80
|
+
});
|
|
81
|
+
|
|
82
|
+
// 设置当前页面
|
|
83
|
+
this.pages.push(this.page);
|
|
84
|
+
await this.setupPage(this.page);
|
|
85
|
+
|
|
86
|
+
return this;
|
|
87
|
+
}
|
|
88
|
+
|
|
89
|
+
/**
|
|
90
|
+
* 设置页面(CDP 拦截器 + 消息绑定)
|
|
91
|
+
*/
|
|
92
|
+
async setupPage(page) {
|
|
93
|
+
try {
|
|
94
|
+
const cdp = await page.context().newCDPSession(page);
|
|
95
|
+
|
|
96
|
+
// 1. 启用 Runtime 域
|
|
97
|
+
await cdp.send('Runtime.enable');
|
|
98
|
+
|
|
99
|
+
// 2. 添加 CDP binding(前端调用此函数,后端接收)
|
|
100
|
+
await cdp.send('Runtime.addBinding', { name: '__deepspider_send__' });
|
|
101
|
+
|
|
102
|
+
// 3. 监听 binding 调用
|
|
103
|
+
cdp.on('Runtime.bindingCalled', (event) => {
|
|
104
|
+
if (event.name === '__deepspider_send__') {
|
|
105
|
+
try {
|
|
106
|
+
const data = JSON.parse(event.payload);
|
|
107
|
+
console.log('[BrowserClient] 收到消息:', data.type);
|
|
108
|
+
if (this.onMessage) {
|
|
109
|
+
this.onMessage(data, page);
|
|
110
|
+
}
|
|
111
|
+
} catch (e) {
|
|
112
|
+
console.error('[BrowserClient] 解析消息失败:', e.message);
|
|
113
|
+
}
|
|
114
|
+
}
|
|
115
|
+
});
|
|
116
|
+
|
|
117
|
+
// 4. 启动拦截器
|
|
118
|
+
const networkInterceptor = new NetworkInterceptor(cdp, page);
|
|
119
|
+
const scriptInterceptor = new ScriptInterceptor(cdp, page);
|
|
120
|
+
await networkInterceptor.start();
|
|
121
|
+
await scriptInterceptor.start();
|
|
122
|
+
|
|
123
|
+
// 保存引用
|
|
124
|
+
if (page === this.page) {
|
|
125
|
+
this.cdpSession = cdp;
|
|
126
|
+
this.networkInterceptor = networkInterceptor;
|
|
127
|
+
this.scriptInterceptor = scriptInterceptor;
|
|
128
|
+
}
|
|
129
|
+
|
|
130
|
+
// 监听页面导航
|
|
131
|
+
page.on('framenavigated', (frame) => {
|
|
132
|
+
if (frame === page.mainFrame()) {
|
|
133
|
+
console.log('[BrowserClient] 页面导航到:', frame.url());
|
|
134
|
+
}
|
|
135
|
+
});
|
|
136
|
+
|
|
137
|
+
console.log('[BrowserClient] 页面已设置:', page.url() || '(空白页)');
|
|
138
|
+
} catch (e) {
|
|
139
|
+
console.error('[BrowserClient] 设置页面失败:', e.message);
|
|
140
|
+
}
|
|
141
|
+
}
|
|
142
|
+
|
|
143
|
+
/**
|
|
144
|
+
* 获取 CDP 会话(始终使用当前页面的 session)
|
|
145
|
+
*/
|
|
146
|
+
async getCDPSession() {
|
|
147
|
+
// 每次都为当前页面创建新的 CDP session,确保上下文正确
|
|
148
|
+
if (this.page) {
|
|
149
|
+
try {
|
|
150
|
+
this.cdpSession = await this.page.context().newCDPSession(this.page);
|
|
151
|
+
} catch (e) {
|
|
152
|
+
console.error('[BrowserClient] 创建 CDP session 失败:', e.message);
|
|
153
|
+
return null;
|
|
154
|
+
}
|
|
155
|
+
}
|
|
156
|
+
return this.cdpSession;
|
|
157
|
+
}
|
|
158
|
+
|
|
159
|
+
/**
|
|
160
|
+
* 导航到 URL
|
|
161
|
+
*/
|
|
162
|
+
async navigate(url, options = {}) {
|
|
163
|
+
const { waitUntil = 'domcontentloaded' } = options;
|
|
164
|
+
await this.page.goto(url, { waitUntil });
|
|
165
|
+
return this.page.url();
|
|
166
|
+
}
|
|
167
|
+
|
|
168
|
+
/**
|
|
169
|
+
* 获取当前页面
|
|
170
|
+
*/
|
|
171
|
+
getPage() {
|
|
172
|
+
return this.page;
|
|
173
|
+
}
|
|
174
|
+
|
|
175
|
+
/**
|
|
176
|
+
* 获取浏览器上下文
|
|
177
|
+
*/
|
|
178
|
+
getContext() {
|
|
179
|
+
return this.context;
|
|
180
|
+
}
|
|
181
|
+
|
|
182
|
+
/**
|
|
183
|
+
* 关闭浏览器
|
|
184
|
+
*/
|
|
185
|
+
async close() {
|
|
186
|
+
if (this.cdpSession) {
|
|
187
|
+
await this.cdpSession.detach().catch(() => {});
|
|
188
|
+
this.cdpSession = null;
|
|
189
|
+
}
|
|
190
|
+
if (this.browser) {
|
|
191
|
+
await this.browser.close();
|
|
192
|
+
this.browser = null;
|
|
193
|
+
this.context = null;
|
|
194
|
+
this.page = null;
|
|
195
|
+
}
|
|
196
|
+
}
|
|
197
|
+
}
|