Assert 的 MCP 服务器 — 让您的 AI 编码助手无需离开聊天界面即可生成、运行和检查 E2E 测试。
用简单的英语描述用户流程。Assert 会生成 Playwright 测试，在真实的浏览器中执行它，并返回步骤级的测试结果和失败截图 — 所有这些都只需在 Cursor、Claude、Windsurf 或任何兼容 MCP 的代理中进行一次工具调用即可完成。
在 assert.click 免费注册 以在使用此包之前获取您的 API 密钥和项目 ID。
生成 — 描述一个流程，获取可保存并运行的 Markdown 场景
运行 — 针对真实的 Chromium 浏览器执行已保存的场景或即时 Markdown
检查 — 获取步骤级的通过/失败状态、错误消息和失败截图 URL
列表 — 浏览项目的已保存场景
Related MCP server: deploytest MCP server
Node.js >=18.17
项目范围的 Assert 密钥 — 在 assert.click 获取
1. 在您的仓库中创建 assert.config.json：
{
"projectApiKey": "assert_project_key_here",
"projectId": "project_123"
}2. 将 MCP 服务器添加到您的代理配置中：
{
"mcpServers": {
"assert": {
"command": "npx",
"args": ["-y", "@assert-click/mcp"],
"env": {
"ASSERT_CONFIG": "/absolute/path/to/assert.config.json"
}
}
}
}就是这样。您的代理现在可以访问所有四个 Assert 工具。
ASSERT_API_KEY: API 密钥（存储在 assert.config.json 中的替代方案）
ASSERT_PROJECT_ID: 可选的默认项目 ID
ASSERT_CONFIG: 可选的配置文件或目录路径
MCP 服务器将从当前目录向上查找这些文件：
assert.config.json
assert.config.local.json
assert.config.local.json 会合并在 assert.config.json 之上。
如果您更喜欢基于环境变量的密钥而不是提交密钥：
{
"projectApiKeyEnv": "ASSERT_API_KEY",
"projectId": "project_123"
}assert_generate根据简单的英语描述生成 Markdown 场景。
输入：
description: string — 用户应该能够做什么
url: string — 起始 URL
project_id?: string
save?: boolean — 保存到项目（默认：true）
返回：
{
"scenario_id": "scenario_123",
"markdown": "URL: https://example.com/login\nSCENARIO: Login\nPROCESS:\n - Fill \"email\" with \"user@example.com\"\nEXPECT: Dashboard",
"saved": true
}assert_run在真实的浏览器中执行已保存的场景或即时 Markdown。
输入：
scenario_id?: string
markdown?: string
project_id?: string
request_id?: string
必须提供 scenario_id 或 markdown 中的一个。
返回：
{
"run_id": "run_123",
"status": "queued",
"estimated_duration_seconds": null
}assert_status获取运行的步骤级结果。
输入：
run_id: string
返回：
{
"run_id": "run_123",
"status": "passed",
"duration_ms": 4200,
"steps": [
{
"description": "Fill email",
"status": "passed",
"error": null,
"screenshot_url": null
}
],
"failure_summary": null,
"full_log_url": null
}assert_list列出项目的已保存场景。
输入：
project_id?: string
cursor?: string
limit?: number
返回：
{
"scenarios": [
{
"id": "scenario_123",
"name": "Login flow",
"project_id": "project_123",
"last_run_status": "passed",
"last_run_at": "2026-03-31T10:00:00.000Z",
"url": "https://example.com/login"
}
],
"next_cursor": null
}错误以结构化 JSON 形式返回：
{
"error": {
"code": "INVALID_API_KEY",
"message": "The ASSERT_API_KEY is invalid or missing.",
"field": null
}
}常见代码：
INVALID_API_KEY
SCENARIO_NOT_FOUND
RUN_NOT_FOUND
VALIDATION_ERROR
UPSTREAM_ERROR
MIT
Generate a ready-to-run E2E test scenario in Assert Markdown format from a plain-English description. Optionally save it to Assert for later execution.
| Name | Required | Description | Default |
|---|---|---|---|
| description | Yes | Plain English description of what to test. | |
| url | Yes | The base URL of the app under test. | |
| project_id | No | Optional. Associate with a project. | |
| save | No | Optional. If true, save the scenario to Assert. Defaults to false — returns markdown preview only. |
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 discloses key behavioral traits: it generates test scenarios, can optionally save them to Assert, and defaults to returning a markdown preview. However, it lacks details on permissions needed, rate limits, error handling, or what 'ready-to-run' entails (e.g., format specifics, dependencies). This is adequate but leaves gaps for a mutation-capable 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 front-loaded with the core purpose in the first sentence, followed by an optional feature in the second. Both sentences earn their place by clarifying functionality and user choice. It is appropriately sized with zero waste or redundancy.
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 no annotations, 4 parameters with 100% schema coverage, and no output schema, the description is moderately complete. It covers the tool's purpose and key behavior (generation and optional saving), but lacks details on output format (beyond 'markdown preview'), error cases, or integration context. For a tool that creates test scenarios, more guidance on output expectations would be helpful.
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 thoroughly. The description adds marginal value by implying the 'description' parameter is for plain-English input and 'save' controls persistence, but does not provide additional syntax, format, or usage details beyond what the schema states. Baseline 3 is appropriate when the schema does the heavy lifting.
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 specific action ('Generate a ready-to-run E2E test scenario'), the resource ('in Assert Markdown format'), and the transformation ('from a plain-English description'). It distinguishes from sibling tools like assert_list, assert_run, and assert_status by focusing on test creation rather than listing, executing, or checking status.
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 clear context for when to use this tool: when you need to create E2E test scenarios from English descriptions. However, it does not explicitly state when not to use it or name alternatives (e.g., use assert_run for execution, assert_list for browsing). The optional save parameter implies a choice between preview and persistence, but no explicit guidance on choosing between this and other tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List existing E2E test scenarios saved in Assert. Use this to audit coverage before generating new tests. Supports filtering by project and pagination.
| Name | Required | Description | Default |
|---|---|---|---|
| project_id | No | Optional. Filter by project ID. | |
| cursor | No | Optional. Pagination cursor from previous response. | |
| limit | No | Optional. Max results to return. Default 20, max 100. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions 'Supports filtering by project and pagination,' which adds useful context beyond basic listing. However, it lacks details on permissions, rate limits, or response format, leaving gaps for a tool with no annotations.
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 two sentences, front-loaded with the core purpose and usage context. Every sentence earns its place by providing essential information without redundancy, making it highly efficient and well-structured.
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 no annotations and no output schema, the description does well by covering purpose, usage guidelines, and behavioral hints like filtering and pagination. However, it lacks details on return values or error handling, which could be useful for a list tool with no structured output information.
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 three parameters thoroughly. The description adds marginal value by mentioning filtering by project and pagination, but does not provide additional syntax or format details beyond what the schema specifies, aligning with the baseline for 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 'List' and resource 'existing E2E test scenarios saved in Assert,' making the purpose specific and unambiguous. It distinguishes from siblings like 'assert_generate' (create), 'assert_run' (execute), and 'assert_status' (check status) by focusing on retrieval of saved test scenarios.
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 explicitly states when to use this tool: 'to audit coverage before generating new tests.' This provides clear context for usage and distinguishes it from alternatives like 'assert_generate' for creating tests, offering practical guidance for the agent.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Execute a test scenario and return a run ID. Accepts either a saved scenario_id or ad-hoc markdown. Runs are async — use assert_status to poll for completion.
| Name | Required | Description | Default |
|---|---|---|---|
| scenario_id | No | ID of a saved scenario to run. | |
| markdown | No | Optional. Run an ad-hoc scenario without saving it first. | |
| project_id | No | Optional. Associate an ad-hoc markdown run with a project. | |
| request_id | No | Optional. Client-generated idempotency key. Retrying with the same ID returns the existing run. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden and does well by disclosing key behavioral traits: the operation is async (not immediate), returns a run ID for tracking, and requires polling with assert_status for completion. It doesn't mention error handling, timeout behavior, or authentication requirements, but covers the essential execution model.
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 perfectly concise with three sentences that each earn their place: states the purpose, explains the input options, and provides critical usage guidance about async behavior. No wasted words, front-loaded with core functionality.
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 (async execution with polling), no annotations, and no output schema, the description does well by explaining the execution model and relationship to assert_status. It could be more complete by mentioning what the run ID represents or error cases, but covers the essential context for proper tool selection and invocation.
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 thoroughly. The description adds minimal value beyond the schema by mentioning the scenario_id/markdown alternatives and the async nature, but doesn't provide additional semantic context about parameter interactions or usage patterns.
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 specific action ('Execute a test scenario'), the resource ('test scenario'), and the outcome ('return a run ID'). It distinguishes from siblings by specifying this tool initiates runs while assert_status polls for completion and assert_list/assert_generate handle other operations.
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 explicitly provides when-to-use guidance: use this tool to start a test run, and use assert_status to poll for completion. It also distinguishes between using saved scenarios (scenario_id) vs. ad-hoc markdown, though it doesn't explicitly mention when to choose between these alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Poll a test run for its current status and step-level results. Returns pass/fail with actionable failure details and screenshot URLs for failed steps.
| Name | Required | Description | Default |
|---|---|---|---|
| run_id | Yes | Run ID returned by assert_run. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden and does well by disclosing key behavioral traits: it's a polling operation (implies repeated calls may be needed), returns pass/fail status, provides actionable failure details, and includes screenshot URLs for failed steps. It doesn't mention rate limits, authentication needs, or error handling.
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 concise sentences with zero waste. First sentence states the action and resource, second sentence details the return values. Every word earns its place and information is front-loaded appropriately.
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 single-parameter tool with no output schema, the description provides good context about what the tool returns (pass/fail status, failure details, screenshot URLs). It could be more complete by specifying the polling interval or whether this is a blocking call, but covers the essential behavior well given the complexity.
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% (run_id is fully documented in schema), so baseline is 3. The description doesn't add any parameter-specific information beyond what the schema already provides about the run_id parameter.
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 with specific verbs ('poll', 'returns') and resources ('test run', 'status', 'step-level results'). It distinguishes from siblings by focusing on status checking rather than generating, listing, or initiating runs.
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 usage after a test run has been initiated (since it requires a run_id from assert_run), but doesn't explicitly state when to use this tool versus alternatives. No explicit guidance on when-not-to-use or direct comparison to siblings is provided.
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.
v1.0.7assert_generateassert_listassert_runassert_statusScored across 4 tools
Each tool has a clearly distinct purpose with no overlap: generate creates tests, list retrieves existing ones, run executes tests, and status checks results. The descriptions explicitly differentiate their functions, making it easy for an agent to select the right tool without confusion.
All tool names follow a consistent 'assert_verb' pattern (assert_generate, assert_list, assert_run, assert_status), using snake_case and clear action verbs. This predictability enhances usability and reduces cognitive load for agents.
With 4 tools, the server is well-scoped for E2E test management, covering the full lifecycle: generate, list, run, and status check. Each tool earns its place without bloat, making the set efficient and focused on its domain.
The toolset provides complete CRUD-like coverage for E2E test scenarios: create (generate), read (list), execute (run), and monitor (status). There are no obvious gaps, as it supports both saved and ad-hoc tests with actionable feedback, enabling seamless agent workflows.
Run, debug and inspect Playwright E2E tests from any AI agent: diagnostics, live DOM, selectors.
Browser-based QA for AI-built software. Test pages with real browsers via agents.
Approved test intent, reviewed Playwright automation and run evidence, inside your editor.
Independent preview-URL QA for coding agents. Playwright heuristics, pass/fail pack.
