karate-graph-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| mcp_healthA | Health probe for MCP connectivity and server state. |
| mcp_versionA | Return server and package version metadata. |
| register_projectB | Register a new Karate project for analysis. |
| delete_projectB | Delete a project from the registry and in-memory cache. |
| clear_all_projectsA | Clear all registered projects from the registry and in-memory cache. Useful for resetting the analyzer state. |
| list_projectsA | List all registered projects and their analysis status. |
| analyze_projectB | Analyze a registered project to build its dependency graph. |
| bulk_analyzeA | Analyze all registered projects at once. |
| query_dependenciesC | Find all components that the specified component depends on. |
| impact_analysisB | Identify all test cases and workflows affected by a change in this component. |
| search_apiB | Search for API endpoints in an analyzed project. |
| get_api_statsC | Get API statistics for an analyzed project. |
| get_page_statsB | Get Page statistics for an analyzed project. |
| search_workflowC | Search for workflows or specific scenarios. |
| search_test_caseB | Search for test cases by Jira ID or name. |
| search_java_usageB | Search Java class/method usage and the test cases that call them. |
| search_js_usageC | Search JavaScript file/function usage and the test cases that call them. |
| search_error_patternA | Search failed nodes by error text, fingerprint, or failed-step pattern. |
| search_reusable_functionA | Search Java/JavaScript source for reusable helper functions before adding new code. |
| get_project_healthB | Get an architectural health report (cycles, orphans, complexity). |
| get_failure_hotspotsB | Identify components (APIs, Workflows) that contribute most to test failures. Returns a sorted list of hotspots based on failure impact score. |
| top_hotspotsC | Preset query: top failure hotspots for a project. |
| unused_componentsC | Preset query: unused components flattened across types. |
| common_usage_mapB | Return reusable components sorted by how many test cases use them. |
| javascript_structure_mapC | Return JavaScript files, exported/helper functions, dependencies, and test usage. |
| similar_common_componentsC | Find common/scenario/action components that share the same dependency shape. |
| change_impact_previewB | Preset query: preview impacted test cases from changed files/components. |
| test_selection_suggestionB | Preset query: suggest smallest high-signal test subset to rerun after change. |
| feature_intent_indexC | Build/search scenario intent summaries from feature files. |
| variable_data_flow_traceB | Trace variables in feature scenarios from definition/source to usage. |
| assertion_mapC | Index status/match/assert steps across feature files. |
| call_read_deep_contextC | Return nested call/read context for selected feature scenarios. |
| ai_feature_context_packC | Build AI-ready feature context: intent, variables, assertions, call/read chain, graph context. |
| feature_behavior_mapC | Build scenario behavior maps for AI: preconditions, actions, expectations. |
| scenario_similarity_mapC | Find similar scenarios based on intent keywords for AI reuse and suggestion. |
| feature_reuse_advisorA | Find duplicate feature steps/flows and return AI-safe refactor suggestions. |
| db_query_indexC | Build/search DB query and DB component index. |
| search_db_usageC | Search DB usage by table/query/operation/host/path keywords. |
| db_data_flow_traceC | Trace DB-related variable/call/assertion flow in selected scenarios. |
| db_assertion_mapC | Index DB-related assertions across feature files. |
| db_impact_previewC | Preview impacted tests from changed DB entities. |
| flaky_riskB | Preset query: test cases with mixed pass/fail history (flaky risk). |
| prioritize_fix_queueB | Preset query: rank failures/components to fix first by impact and risk. |
| record_fixA | Record a successful fix for a component and error pattern. This helps the AI 'learn' how to fix similar issues in the future. |
| get_fix_suggestionsB | Get historical fix suggestions for a component and error pattern. |
| auto_fix_hint_packC | Build a step-by-step auto-fix checklist from smart + historical suggestions. |
| get_failure_historyB | Return execution history, flaky score, and failure fingerprint trend for a node. |
| get_failure_debug_contextA | Build an AI-ready debug pack: failure fingerprint, run history, local dependency graph, related hotspots, source snippet, and fix checklist. |
| get_subgraphB | Extract a local subgraph for AI context. Provides a concise view of a node and its neighbors. |
| query_node_by_metadataA | Search nodes by metadata attributes across all projects. Useful for finding all nodes in a specific 'feature' or 'category'. |
| global_searchB | Search across all nodes in all projects using a global keyword. |
| find_pathB | Find all simple paths between two nodes to analyze traceability. |
| get_component_importanceC | Get nodes sorted by their architectural importance (huyết mạch). |
| get_impact_radiusB | Analyze impact within a specific radius for AI reasoning. Identifies components that depend on the specified node. |
| visualize_projectC | Generate an interactive HTML visualization for a project. |
| merge_projectsB | Merge multiple projects into one global dependency graph. |
| export_graphC | Export the dependency graph to a file format. |
| render_execution_reportC | Generate an execution report visualization (Living Graph) with Pass/Fail/Not Run status. |
| process_reports_folderB | Scan a directory for Karate JSON reports, apply them, and generate a visualization. Returns an AI-distilled summary of failures. |
| compare_projectsA | Compare two projects and generate a diff visualization report (Added/Removed/Modified). |
| get_scan_outputC | Get latest persistent scan output data for a project. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 61 tools
Despite the large number of tools, each has a clearly distinct purpose and description, with specific targets like search_api vs search_test_case, and impact_analysis vs change_impact_preview. Overlaps are minimal and descriptions clarify differences.
All tool names follow a consistent verb_noun pattern with underscores, using lowercase throughout. No mixing of styles such as camelCase or different conventions, making the naming predictable and readable.
With 61 tools, the count is extremely high for an MCP server. Even though the domain is broad, the tool surface is overly large and would benefit from consolidation or modularization, as most servers in this ecosystem have fewer than 20 tools.
The tool set covers the full lifecycle of Karate testing analysis: project registration, dependency graph construction, impact analysis, failure debugging, search, presets, and visualization. There are no obvious gaps in the analytical capabilities offered.