cbrowser 6.3.0 β 6.3.1
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/README.md +57 -4
- package/package.json +15 -19
package/README.md
CHANGED
|
@@ -1,11 +1,66 @@
|
|
|
1
1
|
# CBrowser
|
|
2
2
|
|
|
3
|
-
**The
|
|
3
|
+
**The only browser automation that asks: "Can a real user complete this safely?"**
|
|
4
|
+
|
|
5
|
+
Most AI automation tools ask if a task *can* be completed. CBrowser asks if an **elderly first-timer on mobile** can complete itβand whether the automation should even be allowed to try.
|
|
4
6
|
|
|
5
7
|
[](https://www.npmjs.com/package/cbrowser)
|
|
6
8
|
[](https://opensource.org/licenses/MIT)
|
|
7
9
|
|
|
8
|
-
##
|
|
10
|
+
## Why CBrowser Exists
|
|
11
|
+
|
|
12
|
+
Every AI browser tool now has self-healing selectors and natural language commands. That's table stakes.
|
|
13
|
+
|
|
14
|
+
**CBrowser solves three problems no one else does:**
|
|
15
|
+
|
|
16
|
+
### 1. Constitutional AI Safety (No One Else Has This)
|
|
17
|
+
|
|
18
|
+
Other tools will happily click "Delete All Data" or "Transfer $10,000" if you ask. CBrowser classifies every action by risk:
|
|
19
|
+
|
|
20
|
+
| Zone | Actions | Behavior |
|
|
21
|
+
|------|---------|----------|
|
|
22
|
+
| π’ **Green** | Navigate, read, screenshot | Auto-execute |
|
|
23
|
+
| π‘ **Yellow** | Click buttons, fill forms | Log and proceed |
|
|
24
|
+
| π΄ **Red** | Submit, delete, purchase | **Requires verification** |
|
|
25
|
+
| β¬ **Black** | Bypass auth, inject scripts | **Never executes** |
|
|
26
|
+
|
|
27
|
+
This isn't just guardrailsβit's the only AI browser automation with built-in ethical boundaries.
|
|
28
|
+
|
|
29
|
+
### 2. User Perspective Testing (Not Just "Does It Work?")
|
|
30
|
+
|
|
31
|
+
Other tools test if buttons click. CBrowser tests if **real humans** can use your site:
|
|
32
|
+
|
|
33
|
+
```bash
|
|
34
|
+
npx cbrowser compare-personas \
|
|
35
|
+
--start "https://your-site.com" \
|
|
36
|
+
--goal "Complete checkout" \
|
|
37
|
+
--personas power-user,elderly-user,mobile-user
|
|
38
|
+
```
|
|
39
|
+
|
|
40
|
+
Each persona has realistic human behavior: reaction times, typo rates, mouse jitter, attention patterns. You'll discover that your checkout works fine for developers but fails for 40% of real users.
|
|
41
|
+
|
|
42
|
+
### 3. Claude-Native MCP Integration
|
|
43
|
+
|
|
44
|
+
Built for the Claude ecosystem. Add to Claude Desktop and get browser automation as a native capabilityβno API keys, no external services, no vendor lock-in.
|
|
45
|
+
|
|
46
|
+
---
|
|
47
|
+
|
|
48
|
+
## Feature Comparison
|
|
49
|
+
|
|
50
|
+
| Feature | CBrowser | Skyvern | Browser-Use | Testim |
|
|
51
|
+
|---------|:--------:|:-------:|:-----------:|:------:|
|
|
52
|
+
| Self-healing selectors | β
| β
| β
| β
|
|
|
53
|
+
| Natural language | β
| β
| β
| β οΈ |
|
|
54
|
+
| **Constitutional safety zones** | β
| β | β | β |
|
|
55
|
+
| **Multi-persona comparison** | β
| β | β | β |
|
|
56
|
+
| **Human behavior simulation** | β
| β | β | β |
|
|
57
|
+
| **Claude MCP server** | β
| β | β | β |
|
|
58
|
+
| Flaky test detection | β
| β | β | β
|
|
|
59
|
+
| AI test repair | β
| β | β | β οΈ |
|
|
60
|
+
|
|
61
|
+
---
|
|
62
|
+
|
|
63
|
+
## Also Included (Table Stakes)
|
|
9
64
|
|
|
10
65
|
| Traditional Automation | CBrowser |
|
|
11
66
|
|------------------------|----------|
|
|
@@ -15,8 +70,6 @@
|
|
|
15
70
|
| Manual test assertions | **Natural language assertions** |
|
|
16
71
|
| Scripted tests only | **AI test generation** from page analysis |
|
|
17
72
|
| Stateless between runs | Persistent sessions, cookies, localStorage |
|
|
18
|
-
| No user context | Personas with goals, behaviors, limitations |
|
|
19
|
-
| Standalone tool | **MCP Server** for Claude integration |
|
|
20
73
|
|
|
21
74
|
## Quick Start
|
|
22
75
|
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "cbrowser",
|
|
3
|
-
"version": "6.3.
|
|
4
|
-
"description": "
|
|
3
|
+
"version": "6.3.1",
|
|
4
|
+
"description": "The only browser automation with constitutional AI safety zones and user perspective testing. Tests if real humans can complete tasks safely.",
|
|
5
5
|
"main": "dist/index.js",
|
|
6
6
|
"types": "dist/index.d.ts",
|
|
7
7
|
"bin": {
|
|
@@ -20,25 +20,21 @@
|
|
|
20
20
|
"browser-automation",
|
|
21
21
|
"playwright",
|
|
22
22
|
"ai-testing",
|
|
23
|
-
"web-scraping",
|
|
24
|
-
"e2e-testing",
|
|
25
|
-
"persona-testing",
|
|
26
|
-
"session-management",
|
|
27
23
|
"constitutional-ai",
|
|
28
|
-
"
|
|
29
|
-
"
|
|
30
|
-
"
|
|
31
|
-
"
|
|
32
|
-
"core-web-vitals",
|
|
33
|
-
"har-export",
|
|
34
|
-
"network-mocking",
|
|
35
|
-
"video-recording",
|
|
36
|
-
"self-healing-selectors",
|
|
37
|
-
"smart-retry",
|
|
38
|
-
"natural-language-assertions",
|
|
39
|
-
"test-generation",
|
|
24
|
+
"ai-safety",
|
|
25
|
+
"persona-testing",
|
|
26
|
+
"user-perspective-testing",
|
|
27
|
+
"accessibility-testing",
|
|
40
28
|
"mcp-server",
|
|
41
|
-
"claude-integration"
|
|
29
|
+
"claude-integration",
|
|
30
|
+
"anthropic",
|
|
31
|
+
"self-healing-selectors",
|
|
32
|
+
"flaky-test-detection",
|
|
33
|
+
"ai-test-repair",
|
|
34
|
+
"natural-language-tests",
|
|
35
|
+
"e2e-testing",
|
|
36
|
+
"autonomous-agent",
|
|
37
|
+
"human-behavior-simulation"
|
|
42
38
|
],
|
|
43
39
|
"author": "Wyld Digital",
|
|
44
40
|
"license": "MIT",
|