Этот проект содержит Python MCP-серверы для анализа сбоев тестов, ориентированные на QA, построенные с использованием FastMCP.
mcp-server-demo/
├── FailureAnalysisFromLogs.py
├── README.md
├── logs/
├── pyproject.toml
├── testFailureAnalysis.py
└── uv.lockRelated MCP server: CI Investigator MCP
testFailureAnalysis.py
analyze_test_failure
cluster_failures
detect_flaky_tests
FailureAnalysisFromLogs.py
analyze_test_failure
Читает и классифицирует локальные файлы .log из папки logs/
logs/
Примеры журналов сбоев, используемые MCP-сервером на основе журналов
pyproject.toml
Конфигурация версии Python и зависимостей
uv.lock
Зафиксированные версии зависимостей для воспроизводимой установки
.vscode/mcp.json
Записи MCP-сервера для test-failure-analysis, test-failure-analysis-from-logs и mcp-atlassian
Python 3.11 или новее
uv
Доступ в Интернет для первой установки зависимостей
Из корня репозитория:
cd mcp-server-demo
uv syncЗапустите MCP-сервер на основе входных данных с помощью:
uv run python testFailureAnalysis.pyЗапустите MCP-сервер на основе журналов с помощью:
uv run python FailureAnalysisFromLogs.pyanalyze_test_failureАнализирует неудачный тест, используя имя теста, трассировку стека и журналы, затем возвращает вероятную категорию сбоя и рекомендацию.
analyze_test_failure в FailureAnalysisFromLogs.pyАнализирует локальные файлы .log из папки logs/ и возвращает классификацию сбоя, вероятную первопричину, рекомендацию и важные строки ошибки. Он может анализировать все журналы, конкретное имя теста или точное имя файла журнала.
cluster_failuresГруппирует похожие сбои по сигнатуре трассировки стека, чтобы повторяющиеся шаблоны было легче обнаружить.
detect_flaky_testsПросматривает исторические результаты прохождения/сбоев и идентифицирует тесты, которые демонстрируют нестабильное поведение.
Корень репозитория содержит .vscode/mcp.json, который может использоваться инструментами с поддержкой MCP для настройки локального сервера во время разработки. Он включает записи для test-failure-analysis, test-failure-analysis-from-logs и mcp-atlassian.
Локальная конфигурация MCP включает запись сервера mcp-atlassian для доступа к Jira.
Используйте настроенную запись mcp-atlassian из .vscode/mcp.json в вашем клиенте с поддержкой MCP.
Чтобы запустить его вручную, используйте:
JIRA_URL=<your-jira-url> \
JIRA_USERNAME=<your-jira-username> \
JIRA_API_TOKEN=<your-jira-api-token> \
uvx mcp-atlassianЕсли uv недоступен, установите его и перезапустите терминал.
Если установка зависимостей не удалась, убедитесь, что активен Python 3.11+.
Если сервер не запускается, выполните uv sync снова внутри mcp-server-demo.
Analyze a failed test and suggest likely root cause.
| Name | Required | Description | Default |
|---|---|---|---|
| test_name | Yes | ||
| stack_trace | Yes | ||
| logs | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, and the description only gives a high-level purpose without detailing behavior (e.g., whether it is safe, requires permissions, or modifies state). The description does not compensate for missing 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 very short (one sentence), which is concise but lacks necessary detail. It is front-loaded, but could be more informative without losing conciseness.
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 required parameters, no output schema, and no annotations, the description is severely lacking. It does not explain return values, how inputs are used, or any other context needed 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 0%; the description adds no meaning about the three parameters (test_name, stack_trace, logs). They are just named but not explained.
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: analyzing a failed test and suggesting a root cause. However, it does not explicitly differentiate from siblings like 'cluster_failures' or 'detect_flaky_tests', though the intent is distinct.
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, no prerequisites or exclusions provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Group failures by similar error signature.
Each failure should contain test_name and stack_trace.
| Name | Required | Description | Default |
|---|---|---|---|
| failures | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must cover behavioral traits. It only states grouping by error signature but omits details about output, state modifications, error handling, or performance implications.
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, highly concise, and front-loaded with the core action. Every word adds value without 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?
There is no output schema and no description of return values. The tool clusters failures but doesn't indicate what the output looks like (e.g., clusters with centroid signatures). This is a significant gap for a grouping tool.
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?
The single parameter 'failures' is an array of objects without schema descriptions. The description compensates by specifying that each failure should contain 'test_name' and 'stack_trace', adding meaning beyond the schema.
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 'group' and resource 'failures by similar error signature', and specifies required fields. It distinguishes from sibling tools like analyze_test_failure and detect_flaky_tests, though not explicitly.
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 its siblings. There is no mention of scenarios where clustering is appropriate or when to use alternative tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Detect flaky tests from historical pass/fail data.
Each item should contain test_name and status.
| Name | Required | Description | Default |
|---|---|---|---|
| test_history | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description bears full responsibility. It does not disclose whether the tool is read-only, destructive, or has side effects. It only hints at input formatting (test_name and status) but omits behavior like output structure or failure modes.
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 with two sentences: the first states the tool's purpose, and the second provides an input hint. No unnecessary words; every sentence contributes meaning.
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 simple tool with one array parameter and no output schema, the description covers the core purpose and input requirement. However, it lacks details on output (e.g., what the tool returns) and fails to provide context relative to siblings. It is minimally adequate but not fully complete.
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 0%, but the description adds that each item should contain 'test_name and status', which provides necessary meaning beyond the schema (which only says array of objects with additionalProperties). However, it does not specify data types or expected values, so it adds limited value.
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 action ('Detect'), the resource ('flaky tests'), and the data source ('historical pass/fail data'). It distinguishes from sibling tools like 'analyze_test_failure' which focuses on analyzing a specific failure, and 'cluster_failures' which clusters failures, making the tool's purpose distinct.
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 explicit guidance on when to use this tool versus alternatives ('analyze_test_failure', 'cluster_failures'). It implies usage for historical data but does not set context for selection, leaving the agent to infer.
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. Dates show when Glama detected each change.
v0.1.0analyze_test_failurecluster_failuresdetect_flaky_testsEach tool targets a distinct aspect of test failure analysis: root cause analysis for a single failure, grouping failures by error signature, and detecting flaky tests from history. There is no overlap or ambiguity.
All tool names follow a consistent verb_noun pattern using snake_case (analyze_test_failure, cluster_failures, detect_flaky_tests), making them predictable and easy to understand.
With exactly 3 tools, the server is well-scoped for test failure analysis. This number is sufficient to cover core functionalities without being overwhelming or too sparse.
The tool set covers essential operations for analyzing test failures, clustering, and flakiness detection. However, it lacks tools for data retrieval (e.g., fetching test history) or generating summary reports, which are minor gaps.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Flaky test detection, root cause analysis, and fix suggestions for development teams.
BuildPulse CI test analytics for AI agents — flaky tests, coverage, and CI run history.
Structured analysis API and remote MCP tool for text, JSON records and numeric series.
Test the voice agents you run: scored transcripts, pass/fail verdicts, latency and WER metrics.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/rudrathkr/MCPServerCreationPythonSDK'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
