Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Decide Test MCP generate test cases from my login decision table CSV"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Claude-driven testing workflow that generates test cases from decision tables, provides intelligent guidance for test planning, and generates executable test code.
🤖 Claude-Driven Test Planning: Works with Claude via MCP for intelligent test guidance
📊 Decision Table Parsing: Supports CSV, JSON, and Markdown formats
🎭 Playwright Integration: Generates executable Playwright tests
🔌 API Testing: Creates API test suites with proper authentication
🔧 MCP Server: Integrates seamlessly with Claude Code
📝 TypeScript Support: Generates type-safe test code
💰 Zero Cost: No external API keys required
Related MCP server: MCP Test Case Generator
Build the package:
pnpm install
pnpm buildAdd to Claude Code MCP config (~/.claude-code/mcp.json):
{
"mcpServers": {
"decide-test": {
"command": "node",
"args": ["/absolute/path/to/dist/index.js"]
}
}
}Restart Claude Code
pnpm install
pnpm buildOnce the MCP server is installed, you can use it in Claude Code:
Generate test cases from the decision table at docs/examples/decision-tables/login-decision-table.csvClaude Code will:
Parse the decision table
Explore each test case with AI agents
Generate Playwright test code
Save to tests/e2e/generated/
import {
decisionTableParser,
WebAgent,
testCodeGenerator
} from 'decide-test-mcp';
// 1. Parse decision table
const table = await decisionTableParser.parse(
'docs/examples/decision-tables/login-decision-table.csv'
);
// 2. Get guidance for test planning (you provide the steps)
const webAgent = new WebAgent();
const testSteps = [];
for (const testCase of table.test_cases) {
// Get guidance (example steps and recommendations)
const guidance = webAgent.getExplorationGuidance({
url: 'http://localhost:3000',
test_case: testCase,
objective: testCase.name,
});
console.log(guidance.suggested_approach);
console.log('Example steps:', guidance.example_steps);
// You define the actual test steps based on guidance
const steps = [
{ action: 'navigate', target: 'http://localhost:3000/login', description: 'Go to login' },
{ action: 'fill', selector: 'input[name="email"]', value: 'test@example.com', description: 'Enter email' },
{ action: 'click', selector: 'button[type="submit"]', description: 'Click login' },
];
testSteps.push({
test_case_id: testCase.id,
type: 'web',
steps,
});
}
// 3. Generate test code
const generated = await testCodeGenerator.generate({
test_cases: table.test_cases,
steps: testSteps,
framework: 'playwright',
output_path: 'tests/e2e/generated/',
language: 'typescript',
});
console.log(`Generated ${generated.files_generated.length} test files`);Parse a decision table and generate test case specifications.
Example:
{
"table_path": "docs/examples/decision-tables/login-decision-table.csv",
"format": "csv"
}Get guidance and example steps for planning web tests. Claude uses this to understand what test steps to create.
Example:
{
"url": "http://localhost:3000",
"test_case": {...},
"objective": "Login with valid credentials"
}Returns: Suggested approach, example steps, and guidance for Claude to plan the actual test steps.
Execute predefined web test steps using Playwright.
Example:
{
"url": "http://localhost:3000",
"test_case": {...},
"objective": "Login with valid credentials",
"steps": [
{ "action": "navigate", "target": "http://localhost:3000/login", "description": "Go to login" },
{ "action": "fill", "selector": "input[name='email']", "value": "test@example.com", "description": "Enter email" },
{ "action": "click", "selector": "button[type='submit']", "description": "Click login" }
],
"headless": true,
"screenshot_dir": "./screenshots"
}Get guidance and example steps for planning API tests.
Example:
{
"base_url": "http://localhost:3000/api",
"test_case": {...},
"objective": "Create trip via API",
"auth": {
"type": "bearer",
"credentials": {"token": "..."}
}
}Returns: Suggested approach, example API steps, and guidance for Claude to plan the actual API test steps.
Execute predefined API test steps.
Example:
{
"base_url": "http://localhost:3000/api",
"test_case": {...},
"objective": "Create trip via API",
"steps": [
{ "method": "POST", "endpoint": "/auth/login", "body": {...}, "expected_status": 200 },
{ "method": "POST", "endpoint": "/trips", "body": {...}, "expected_status": 201 }
],
"auth": {
"type": "bearer"
}
}Generate executable test code from test cases and steps.
Example:
{
"test_cases": [...],
"steps": [...],
"framework": "playwright",
"output_path": "tests/e2e/generated/",
"language": "typescript"
}Execute generated tests and return results.
Example:
{
"test_path": "tests/e2e/generated/login.spec.ts",
"framework": "playwright",
"reporter": "list"
}Email,Password,Action,Expected Result,Priority
valid@example.com,ValidPass123,Click Login,Login successful,high
invalid@example.com,ValidPass123,Click Login,Show error message,medium{
"feature": "User Login",
"rules": [
{
"id": "TC001",
"conditions": {
"email": "valid",
"password": "valid"
},
"actions": ["click_login"],
"expected": ["redirect_to_dashboard"]
}
]
}# User Login
| Email | Password | Action | Expected Result |
|-------|----------|--------|----------------|
| valid | valid | Click Login | Login successful |
| invalid | valid | Click Login | Show error |See docs/examples/decision-tables/ for complete examples:
login-decision-table.csv - User authentication tests
trip-creation-decision-table.json - Trip creation with tier limits
collaboration-decision-table.md - Collaboration & permissions
# Install dependencies
pnpm install
# Build
pnpm build
# Run in development mode
pnpm dev
# Run tests
pnpm test┌─────────────────────────────────────┐
│ MCP Server │
│ (Model Context Protocol) │
└─────────────────────────────────────┘
│
┌─────────┼─────────┐
│ │ │
▼ ▼ ▼
┌────────┐ ┌──────┐ ┌──────────┐
│ Parser │ │Agents│ │Generator │
└────────┘ └──────┘ └──────────┘Check MCP config path is correct
Verify Node.js is accessible
Check server logs: ~/.claude-code/logs/mcp-ai-testing.log
Restart Claude Code
Check application is running at specified URL
Review test steps for correctness
Try with headless: false to see browser in action
Check selector specificity
Ensure test cases and steps are complete
Check output directory permissions
Review generated code for syntax errors
MIT
For issues and questions:
Documentation: docs/AI_TESTING_WORKFLOW.md
Execute predefined API test steps. Takes a test case and specific API steps to execute.
| Name | Required | Description | Default |
|---|---|---|---|
| auth | No | Authentication configuration | |
| steps | Yes | Array of API test steps to execute | |
| base_url | Yes | Base URL of the API | |
| objective | Yes | Testing objective | |
| test_case | Yes | Test case being executed |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations; description lacks disclosure of side effects (e.g., whether tests modify data), authentication needs, or error handling. Minimal beyond the action itself.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, no superfluous information. Could be more structured but acceptable.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With 5 parameters (4 required, nested objects) and no output schema, description is too brief. Fails to explain relationship between parameters or execution semantics.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% but parameters like 'auth' and 'test_case' are only generically described. Description adds no extra meaning beyond 'takes a test case and steps'.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Clear verb+resource: 'execute predefined API test steps'. Distinguishes from sibling 'execute_web_test' which implies web-based tests, though no explicit differentiation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use vs. siblings like 'generate_test_code' or 'run_generated_tests'. No context on prerequisites or exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Execute predefined web test steps using Playwright. Takes a test case and specific steps to execute.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Base URL of the web application | |
| steps | Yes | Array of test steps to execute | |
| headless | No | Run browser in headless mode (default: true) | |
| objective | Yes | Testing objective | |
| test_case | Yes | Test case being executed | |
| screenshot_dir | No | Directory to save screenshots (optional) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so description carries full burden. Only states it executes steps using Playwright, but does not disclose side effects, failure behavior, authentication needs, or browser visibility (headless mode is in schema but not description).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, concise and front-loaded. Every sentence provides core information without unnecessary detail.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Despite having 6 parameters (4 required) and no output schema, the description is too brief. It does not explain return values, error handling, prerequisites, or test execution environment. More context needed for completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% (all parameters documented). The description adds little beyond schema: it mentions test case and steps but does not elaborate on meaning or usage. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it executes predefined web test steps using Playwright, specifying the inputs (test case and steps). It distinguishes from sibling tools like execute_api_test and generate_test_code.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this tool versus alternatives like execute_api_test or get_web_test_guidance. The description does not mention exclusions or prerequisites.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generate executable test code (Playwright or API tests) from test cases and steps
| Name | Required | Description | Default |
|---|---|---|---|
| steps | Yes | Array of test steps for each test case | |
| style | No | Test code style (default: standard) | |
| language | No | Programming language (default: typescript) | |
| framework | Yes | Test framework to use | |
| test_cases | Yes | Array of test cases | |
| output_path | Yes | Output directory for generated test files |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It only states 'generate executable test code' but does not disclose important behaviors like overwriting files, error handling, permissions, or output format. The description is too minimal for a code generation tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that is front-loaded and contains no fluff. It conveys the essential purpose efficiently.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description is sparse for a complex tool with 6 parameters. It does not explain how parameters like 'style' or 'language' affect output, nor does it provide context on the generated code's structure or next steps. Sibling tools suggest a workflow, but the description lacks that context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters. The description adds no additional meaning beyond what the schema provides. Baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it generates executable test code from test cases and steps, specifying Playwright or API tests. It distinguishes from sibling tools like execute_api_test (which runs tests) and get_api_test_guidance (which provides guidance).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies use when test cases and steps are available, but does not explicitly state when to use this tool versus siblings like run_generated_tests or parse_decision_table. No exclusions or prerequisites are mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get guidance and example steps for planning API tests. Returns structured information to help Claude plan API test steps.
| Name | Required | Description | Default |
|---|---|---|---|
| auth | No | Authentication configuration | |
| base_url | Yes | Base URL of the API | |
| objective | Yes | Testing objective (e.g., "Create trip via API") | |
| test_case | Yes | Test case to explore |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden. It indicates a non-mutating read operation (returns guidance), but does not disclose any behavioral traits such as error conditions, rate limits, or dependencies beyond what is obvious.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise, consisting of two sentences: the first states the action and resource, the second describes the return value. No unnecessary information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the lack of an output schema and the presence of nested object parameters, the description is too brief. It does not explain the structure of the returned guidance or provide context on how to use the output effectively.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3. The description adds no additional meaning to parameters; it merely states general purpose without linking to specific parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly identifies the tool as providing guidance and example steps for planning API tests, which distinguishes it from siblings like 'execute_api_test' (execution) and 'get_web_test_guidance' (web tests).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description does not provide any guidance on when to use this tool versus alternatives. For example, it doesn't specify prerequisites or that it should be used before test execution, nor does it differentiate from other guidance tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get guidance and example steps for planning web tests. Returns structured information to help Claude plan test steps for a web application flow.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Base URL of the web application | |
| objective | Yes | Testing objective (e.g., "Login with valid credentials") | |
| test_case | Yes | Test case to explore |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It does not mention side effects, permissions, rate limits, or output behavior beyond 'returns structured information,' which is insufficient for a safe invocation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, front-loaded with the core purpose, no redundant or tangential information. Every sentence adds value, making it optimally concise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (3 params, nested object, no output schema), the description is too sparse. It fails to explain the return format, how guidance is structured, or how to interpret the output. This leaves the agent with significant ambiguity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% (all parameters have descriptions). The tool description adds no additional parameter information, so it stays at the baseline of 3. The description neither enhances nor detracts from schema semantics.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Get guidance and example steps for planning web tests.' It uses a specific verb-resource pair and distinguishes from sibling tools like 'get_api_test_guidance' (API-focused) and 'execute_web_test' (execution vs. planning).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance on when to use this tool versus alternatives. While siblings are listed, the description does not mention conditions for use or exclusion, leaving the agent to infer context from tool names alone.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Parse a decision table from CSV, JSON, or Markdown format and generate test case specifications
| Name | Required | Description | Default |
|---|---|---|---|
| format | No | Format of the decision table (auto-detected if not specified) | |
| table_path | Yes | Path to the decision table file |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must fully disclose behavioral traits. It states parsing and generation but does not mention limitations, error handling, or that the operation is non-destructive. The auto-detection of format is implied only through the parameter description, not the tool description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence of 16 words, efficiently stating the core function. It is front-loaded with the verb and resource, with no unnecessary words or repetition.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with no output schema and no annotations, the description is minimally complete for a simple parser. However, it lacks context on the output format and how it integrates with sibling tools, which are all test-related. The 100% schema coverage covers parameters, but the overall tool context is only partially addressed.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with descriptions for both parameters. The tool description does not add additional meaning beyond what is already in the schema (e.g., the formats listed are already in the enum). Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly specifies the action ('Parse'), the resource ('decision table from CSV, JSON, or Markdown format'), and the output ('generate test case specifications'). It effectively distinguishes this tool from siblings, none of which mention parsing or decision tables.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives, no prerequisites, and no indication that it is a preparatory step for other tools. Sibling tools like execute_api_test suggest a workflow, but the description does not connect them.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Execute generated tests and return results
| Name | Required | Description | Default |
|---|---|---|---|
| reporter | No | Test reporter format (default: list) | |
| framework | Yes | Test framework | |
| test_path | Yes | Path to the test file to execute |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It only says 'Execute generated tests and return results' without disclosing behavioral traits like required prior steps (e.g., test generation), failure behavior, or side effects. This is insufficient for a test execution tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single short sentence with no waste, but it is under-specified. It lacks detail expected for a tool with multiple parameters and siblings, making it minimally informative rather than appropriately concise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With 3 parameters, no output schema, and sibling tools that are more specific, the description is too brief. It does not explain what 'return results' entails (e.g., pass/fail, logs), nor does it clarify how this tool relates to generate_test_code. The description is incomplete for effective use.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with descriptions for all three parameters. The description does not add any parameter-specific meaning beyond what's in the schema. For example, it doesn't clarify expected format for test_path or allowed values beyond enums. Baseline 3 is appropriate given high coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Execute' and resource 'generated tests', indicating it runs tests. However, it does not distinguish from sibling tools like execute_api_test or execute_web_test, which are more specific. The purpose is clear but lacks differentiation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. There is no mention of prerequisites, exclusions, or context for selection among siblings like execute_api_test or execute_web_test.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0execute_api_testexecute_web_testgenerate_test_codeget_api_test_guidanceget_web_test_guidanceparse_decision_tablerun_generated_testsScored across 7 tools
Each tool targets a specific function: API vs web execution, guidance, parsing, code generation, and test execution. No two tools have overlapping purposes, ensuring clear selection.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., execute_api_test, get_web_test_guidance). The naming is predictable and uniform throughout.
With 7 tools, the server covers the essential testing workflow without being too sparse or bloated. Each tool serves a necessary role in planning, generating, or executing tests.
The tool set covers the main lifecycle: guidance for planning, parsing decision tables, generating code, and executing tests. Minor gaps like persistent test case storage or advanced reporting exist but are not critical for the core workflow.
Direct access to Cypress tests results and accessibility reports in your AI workflow.
3rd Generation Testing (3TG) — generate deterministic test suites from Markdown spec tables via MCP.
Teamfight Tactics data & AI coaching for Claude and ChatGPT — 19 tools, built-in Riot key.
Generates unit tests for Python code with coverage before/after reports and concrete edge cases.
