Bug, Defect & Advanced Testing

MCP tool reference for defect management, performance and security testing, code export, and AI-powered impact analysis in ContextQA.

Who is this for? SDETs, developers, and DevOps engineers integrating ContextQA with AI coding assistants (Claude, Cursor) or CI/CD pipelines.

This reference covers three capability groups: defect lifecycle tools that bridge test failures to your ALM, advanced testing tools for performance and security, and AI analysis tools that surface test impact and code context.

---

Bug & Defect

---

create_defect_ticket

Compiles failure evidence from an execution and pushes a structured defect ticket to the connected ALM (Jira or Azure DevOps); use this immediately after a failed execution to file a bug without leaving your AI assistant.

Category: Bug & Defect | Authentication required: Yes

Parameters

Name	Required	Type	Description
execution_id	✅	string	ID of the failed test execution to report.
project_id	✅	string	ALM project identifier to create the ticket in.

Returns

JSON with the created ticket's id, url, title, and the ALM system it was filed against.

Example

{
  "execution_id": "exec_4d8f1a",
  "project_id": "CQA"
}

Related Tools

get_test_case_results, investigate_failure, get_root_cause

---

get_auto_healing_suggestions

Retrieves AI-generated proposals for repairing broken locators detected in a failed execution; review suggestions before applying them with approve_auto_healing.

Category: Bug & Defect | Authentication required: Yes

Parameters

Name	Required	Type	Description
execution_id	✅	string	ID of the execution that produced broken locators.

Returns

JSON array of healing proposals. Each item includes healing_id, the original broken locator, the proposed replacement locator, a confidence score, and the affected test step.

Example

{
  "execution_id": "exec_4d8f1a"
}

Related Tools

approve_auto_healing, get_test_step_results, investigate_failure

---

approve_auto_healing

Applies a specific auto-healing patch to update the broken locator in the test case definition; always review the suggestion from get_auto_healing_suggestions first.

Category: Bug & Defect | Authentication required: Yes

Parameters

Name	Required	Type	Description
healing_id	✅	string	ID of the healing proposal to apply. Obtained from get_auto_healing_suggestions.
execution_id	❌	string	Execution context for the heal. Helps the engine track which run triggered the repair.

Returns

JSON confirmation with status, the updated locator value, and the test step that was modified.

Example

{
  "healing_id": "heal_7c3b2e",
  "execution_id": "exec_4d8f1a"
}

Related Tools

get_auto_healing_suggestions, get_test_case_steps, update_test_case_step

---

Advanced Testing

---

execute_performance_test

Runs a load test against the flows covered by a test case using a configurable virtual user count, duration, and ramp-up period.

Category: Advanced Testing | Authentication required: Yes

Parameters

Name	Required	Type	Description
test_case_id	✅	integer	Numeric ID of the test case to use as the load scenario.
virtual_users	✅	integer	Number of concurrent virtual users to simulate.
duration	✅	integer	Total test duration in seconds.
ramp_up	✅	integer	Time in seconds to ramp up to the full virtual user count.

Returns

JSON with a performance execution ID, configuration echo, and a link to the live results dashboard.

Example

{
  "test_case_id": 88,
  "virtual_users": 100,
  "duration": 300,
  "ramp_up": 60
}

Related Tools

get_test_cases, get_execution_status, execute_test_case

---

execute_security_dast_scan

Launches a DAST (Dynamic Application Security Testing) scan against the application exercised by a test case, using the specified scan profile to control scan depth and aggression.

Category: Advanced Testing | Authentication required: Yes

Parameters

Name	Required	Type	Description
test_case_id	✅	integer	Numeric ID of the test case whose flows define the scan target.
scan_profile	✅	string	Scan intensity profile. Accepted values: "standard", "aggressive". Use "standard" for CI pipelines; "aggressive" for dedicated security runs.

Returns

JSON with a DAST scan execution ID, scan profile confirmation, and a link to the vulnerability report once the scan completes.

Example

{
  "test_case_id": 88,
  "scan_profile": "standard"
}

Related Tools

get_test_cases, get_execution_status, execute_test_case

---

export_test_case_as_code

Exports one or more test cases as runnable code in a specified framework and language, writing the output to a local path.

Category: Advanced Testing | Authentication required: Yes

Parameters

Name	Required	Type	Description
test_case_id	✅	integer	Numeric ID of the test case to export.
framework_type	✅	string	Target framework, e.g. "playwright", "cypress", "selenium".
language	✅	string	Output language, e.g. "typescript", "javascript", "python".
destination_path	✅	string	Absolute local path where the generated file(s) should be written.

Returns

JSON confirmation with the destination_path, a download_url to retrieve the file from ContextQA, and the list of generated files.

Example

{
  "test_case_id": 55,
  "framework_type": "playwright",
  "language": "typescript",
  "destination_path": "/home/dev/projects/my-app/tests/generated"
}

Related Tools

export_to_playwright, get_test_cases, migrate_repo_to_contextqa

---

AI-Powered Analysis

---

get_root_cause

Sends failed step data, screenshots, and the session recording to the ContextQA root cause engine and returns a structured AI root cause analysis; use this as the first step in any failure investigation.

Category: AI Analysis | Authentication required: Yes

Parameters

Name	Required	Type	Description
execution_id	✅	string	ID of the failed execution to analyse.

Returns

JSON with rootCause (a plain-English summary), affectedStep, errorType, evidenceLinks (screenshots, recording), and suggested remediation.

Example

{
  "execution_id": "exec_4d8f1a"
}

Related Tools

investigate_failure, get_ai_reasoning, get_test_step_results, create_defect_ticket

---

query_repository

Queries a connected source code repository with a natural-language question and returns the most relevant code context and file references; use this to understand how the application under test is implemented before writing or debugging tests.

Category: AI Analysis | Authentication required: Yes

Parameters

Name	Required	Type	Description
question	✅	string	Natural-language question about the codebase.
repo_url	✅	string	URL of the Git repository to query (HTTPS or SSH).
limit_files	❌	integer	Maximum number of file references to return. Defaults to 10.

Returns

JSON with a answer string, a files array of relevant file paths with matched snippets, and a confidence score.

Example

{
  "question": "How does the checkout flow calculate shipping costs?",
  "repo_url": "https://github.com/example-org/storefront",
  "limit_files": 5
}

Related Tools

analyze_test_impact, generate_tests_from_code_change, analyze_test_repo

---

analyze_test_impact

Determines which existing test cases are affected by a set of changed files (and optionally changed functions), and returns a prioritised list of tests to rerun; use this in CI after a PR diff is known.

Category: AI Analysis | Authentication required: Yes

Parameters

Name	Required	Type	Description
changed_files	✅	array of strings	List of file paths that changed in the current diff, relative to repo root.
changed_functions	❌	array of strings	List of function or method names that changed. Improves precision of impact analysis.

Returns

JSON with affectedTestCases (array of test case objects with IDs and names) and recommendedRerunIds (array of integers ready to pass to execute_test_case).

Example

{
  "changed_files": [
    "src/checkout/ShippingCalculator.ts",
    "src/cart/CartSummary.tsx"
  ],
  "changed_functions": [
    "calculateShipping",
    "applyDiscount"
  ]
}

Related Tools

query_repository, generate_tests_from_code_change, execute_test_case, get_test_cases

---