abtestresult-mcp 1.0.8

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.
Files changed (4) hide show
  1. package/LICENSE +67 -0
  2. package/README.md +76 -0
  3. package/dist/index.js +35224 -0
  4. package/package.json +42 -0
package/LICENSE ADDED
@@ -0,0 +1,67 @@
1
+ ABTestResult MCP Server Personal Use License v1.0
2
+
3
+ Copyright (c) 2026 ABTestResult. All rights reserved.
4
+
5
+ This software is source-available, not open source. You may read it, install it,
6
+ run it, and modify it only under the terms below.
7
+
8
+ 1. Permitted Use
9
+
10
+ You are granted a limited, revocable, non-exclusive, non-transferable,
11
+ non-sublicensable license to use, run, and modify this software solely for your
12
+ own individual, personal, non-commercial use on systems that you own or control.
13
+ Local personal experimentation, learning, and evaluation are permitted.
14
+
15
+ 2. Commercial Use Prohibited
16
+
17
+ You may not use this software, directly or indirectly, for any commercial
18
+ purpose without a separate written commercial license from ABTestResult.
19
+ Commercial use includes, but is not limited to:
20
+
21
+ - selling, licensing, sublicensing, renting, or redistributing the software
22
+ - providing the software or its output as part of a paid or revenue-generating
23
+ product, service, subscription, consulting engagement, or client deliverable
24
+ - using the software in connection with a business, employer, agency, or other
25
+ organization, whether for internal operations or external services
26
+ - using the software to support advertising, lead generation, affiliate income,
27
+ sponsored work, or any other activity intended to generate money or
28
+ commercial advantage
29
+
30
+ 3. Distribution Prohibited
31
+
32
+ You may not publish, distribute, sublicense, sell, transfer, or make this
33
+ software available to any third party, in source or compiled form, without
34
+ prior written permission from ABTestResult.
35
+
36
+ 4. Hosted and Shared Use Prohibited
37
+
38
+ You may not host this software for others, expose it as a network service,
39
+ bundle it into another tool, or otherwise make it available for team, company,
40
+ client, public, or shared use without a separate written commercial license.
41
+
42
+ 5. Modification Scope
43
+
44
+ You may modify the software only for your own permitted personal use. You may
45
+ not use modified versions for any prohibited purpose, and you may not
46
+ distribute modified versions without prior written permission from ABTestResult.
47
+
48
+ 6. Attribution and Notices
49
+
50
+ You must retain this copyright notice and this license in all copies you make
51
+ for your own permitted use.
52
+
53
+ 7. Commercial Licensing
54
+
55
+ If you want to use this software in any commercial setting or money-making
56
+ activity, you must obtain a separate commercial license from ABTestResult.
57
+ Contact: hello@abtestresult.com
58
+
59
+ 8. No Warranty
60
+
61
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
62
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
63
+ FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT. IN NO EVENT SHALL THE
64
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES, OR OTHER
65
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT, OR OTHERWISE, ARISING FROM,
66
+ OUT OF, OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
67
+ SOFTWARE.
package/README.md ADDED
@@ -0,0 +1,76 @@
1
+ # ABTestResult MCP Server
2
+
3
+ Statistical tools for A/B testing, available as a [Model Context Protocol](https://modelcontextprotocol.io) (MCP) server. Powered by [abtestresult.com](https://abtestresult.com).
4
+
5
+ Give any AI assistant (Claude, Cursor, VS Code, etc.) the ability to analyze A/B tests, calculate sample sizes, run Bayesian analysis, and more. All calculations run locally on your machine — no API keys, your test data never leaves your device.
6
+
7
+ This package ships readable source. It is source-available for local personal use only and is not licensed for commercial use.
8
+
9
+ ## Setup
10
+
11
+ Add to your MCP configuration:
12
+
13
+ ```json
14
+ {
15
+ "mcpServers": {
16
+ "abtestresult": {
17
+ "command": "npx",
18
+ "args": ["-y", "abtestresult-mcp"]
19
+ }
20
+ }
21
+ }
22
+ ```
23
+
24
+ **Where to add this:**
25
+
26
+ - **Claude Desktop:** `~/.claude/claude_desktop_config.json`
27
+ - **Claude Code:** `.mcp.json` in your project root
28
+ - **Cursor:** `.cursor/mcp.json`
29
+ - **VS Code:** `.vscode/settings.json` (use `mcp.servers` instead of `mcpServers`)
30
+
31
+ That's it. No API keys, no authentication. The server starts automatically when your AI assistant needs it.
32
+
33
+ ## Tools
34
+
35
+ | Tool | Description |
36
+ |------|-------------|
37
+ | `analyze_ab_test` | Z-test for rate metrics (conversion rate, CTR, etc.) |
38
+ | `analyze_ab_test_average` | T-test for continuous metrics (revenue, time on page, etc.) |
39
+ | `calculate_sample_size` | Required sample size for rate or average metrics |
40
+ | `calculate_mde` | Minimum detectable effect given a fixed sample size |
41
+ | `bayesian_ab_test` | Bayesian analysis with probability to be best & expected loss |
42
+ | `check_srm` | Sample Ratio Mismatch detection |
43
+ | `paired_test` | Paired T-test or Wilcoxon signed-rank for before/after data |
44
+ | `survey_sample_size` | Cochran's formula with finite population correction |
45
+
46
+ ## Usage Examples
47
+
48
+ Once connected, just ask your AI assistant in natural language:
49
+
50
+ - "My A/B test had 5,000 users per group. Control converted at 5%, variant at 6%. Is it significant?"
51
+ - "How many users do I need to detect a 10% lift on a 3% conversion rate?"
52
+ - "Run a Bayesian analysis on my test: control 10,000 users / 500 conversions, variant 10,000 users / 580 conversions"
53
+ - "Check if my traffic split is valid: control has 10,234 users, variant has 9,766"
54
+ - "What's the smallest effect I can detect with 50,000 total users and a 5% baseline?"
55
+
56
+ ## Privacy
57
+
58
+ Your test data never leaves your machine — all calculations run locally. We collect anonymous usage telemetry (tool name and client type only) to monitor performance and guide development. No test inputs, results, or user identifiers are tracked.
59
+
60
+ ## Learn More
61
+
62
+ - [Full documentation](https://abtestresult.com/mcp-server) — setup guides, tool reference, and rate limits
63
+ - [Web calculators](https://abtestresult.com) — use the same tools in your browser
64
+
65
+ ## License
66
+
67
+ Copyright (c) 2026 ABTestResult. All rights reserved.
68
+
69
+ This package is source-available, not open source.
70
+
71
+ You may use and modify it only for your own local personal, non-commercial use.
72
+ Any commercial use, business use, client use, hosted use, redistribution, or
73
+ money-making activity requires a separate written license from ABTestResult:
74
+ hello@abtestresult.com.
75
+
76
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND.