imperium-crawl 2.3.0 → 2.3.1
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/.env.example +49 -0
- package/dist/constants.d.ts +1 -1
- package/dist/constants.js +1 -1
- package/package.json +3 -2
package/.env.example
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
# ── API Keys ──────────────────────────────────────────────
|
|
2
|
+
# Brave Search — unlocks 4 search tools (web, news, image, video)
|
|
3
|
+
# Free tier: https://brave.com/search/api/
|
|
4
|
+
BRAVE_API_KEY=
|
|
5
|
+
|
|
6
|
+
# LLM — unlocks ai_extract tool + llm_fallback in extract
|
|
7
|
+
# Supports Anthropic, OpenAI, or MiniMax API keys
|
|
8
|
+
LLM_API_KEY=
|
|
9
|
+
LLM_PROVIDER=anthropic # anthropic | openai | minimax
|
|
10
|
+
LLM_MODEL= # override default model (optional)
|
|
11
|
+
|
|
12
|
+
# OpenAI — Whisper transcription for YouTube videos without captions
|
|
13
|
+
# https://platform.openai.com/
|
|
14
|
+
OPENAI_API_KEY=
|
|
15
|
+
|
|
16
|
+
# 2Captcha — auto CAPTCHA solving (reCAPTCHA v2/v3, hCaptcha, Turnstile)
|
|
17
|
+
# https://2captcha.com/
|
|
18
|
+
TWOCAPTCHA_API_KEY=
|
|
19
|
+
|
|
20
|
+
# ── Proxy ─────────────────────────────────────────────────
|
|
21
|
+
# Single proxy (http/https/socks4/socks5)
|
|
22
|
+
PROXY_URL=
|
|
23
|
+
# Rotating proxy pool (comma-separated)
|
|
24
|
+
PROXY_URLS=
|
|
25
|
+
|
|
26
|
+
# ── Browser ───────────────────────────────────────────────
|
|
27
|
+
# Max pooled browser instances (default: 3)
|
|
28
|
+
BROWSER_POOL_SIZE=3
|
|
29
|
+
# Chrome user data dir for authenticated sessions
|
|
30
|
+
CHROME_PROFILE_PATH=
|
|
31
|
+
|
|
32
|
+
# ── Security ──────────────────────────────────────────────
|
|
33
|
+
# 32-byte hex key for encrypting session files at rest
|
|
34
|
+
# Generate: node -e "console.log(require('crypto').randomBytes(32).toString('hex'))"
|
|
35
|
+
SESSION_ENCRYPTION_KEY=
|
|
36
|
+
|
|
37
|
+
# ── Behavior ──────────────────────────────────────────────
|
|
38
|
+
# Honor robots.txt (default: true)
|
|
39
|
+
RESPECT_ROBOTS=true
|
|
40
|
+
|
|
41
|
+
# ── Instagram (advanced — for influencer recipes) ─────────
|
|
42
|
+
# IG_SESSION_ID=
|
|
43
|
+
# IG_CSRF_TOKEN=
|
|
44
|
+
# IG_DS_USER_ID=
|
|
45
|
+
|
|
46
|
+
# ── Debug ─────────────────────────────────────────────────
|
|
47
|
+
# DEBUG=1 # enable debug logging to stderr
|
|
48
|
+
# VERBOSE=1 # same as DEBUG=1
|
|
49
|
+
# NO_COLOR=1 # disable colored output
|
package/dist/constants.d.ts
CHANGED
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
export declare const PACKAGE_NAME = "imperium-crawl";
|
|
2
|
-
export declare const PACKAGE_VERSION = "2.3.
|
|
2
|
+
export declare const PACKAGE_VERSION = "2.3.1";
|
|
3
3
|
export declare const DEFAULT_TIMEOUT_MS = 30000;
|
|
4
4
|
export declare const DEFAULT_MAX_PAGES = 10;
|
|
5
5
|
export declare const DEFAULT_MAX_DEPTH = 2;
|
package/dist/constants.js
CHANGED
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "imperium-crawl",
|
|
3
|
-
"version": "2.3.
|
|
3
|
+
"version": "2.3.1",
|
|
4
4
|
"description": "Open-source CLI tool for web scraping, crawling, search, and custom skills",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"bin": {
|
|
@@ -14,7 +14,8 @@
|
|
|
14
14
|
"./stealth/*": "./dist/stealth/*.js"
|
|
15
15
|
},
|
|
16
16
|
"files": [
|
|
17
|
-
"dist"
|
|
17
|
+
"dist",
|
|
18
|
+
".env.example"
|
|
18
19
|
],
|
|
19
20
|
"scripts": {
|
|
20
21
|
"build": "tsc && shx chmod +x dist/index.js",
|