TeamCity MCP Server

# TeamCity MCP Tools & Capabilities Guide ## Overview The TeamCity MCP (Model Context Protocol) server provides a set of tools for interacting with JetBrains TeamCity CI/CD server via AI-powered coding assistants using MCP tools. This document provides a reference for the available tools, their capabilities, and practical workflows. > For exhaustive argument and mode details, see the [MCP Tool Reference](./mcp-tools-reference.md). This guide focuses on practical usage patterns and stays aligned with the shipped tool surface. ## Architecture Overview ### Core Components 1. **MCP Server** - Handles protocol communication and tool execution 2. **TeamCity API Client** - Singleton client for REST access with error handling 3. **Tool Registry** - Dynamic tool loading based on operation mode 4. **Pagination Utilities** - Consistent list pagination helpers 5. **Error Handling & Logging** - Structured errors; logs redact sensitive values ### Key Features - **Two Operation Modes**: Dev (safe) and Full (modifying resources) - **Type Safety**: TypeScript with Zod input validation - **Pagination Support**: Standardized list pagination across tools - **Error Handling**: Consistent error shaping; logs redact sensitive values ## Current Implementation Notes The live implementation emphasizes a simple, direct architecture. A few practical details to help you use the tools effectively: - Response shape: List/get tools return a JSON object encoded as text in the MCP `content` field. For lists, the payload is typically: `{ "items": [...], "pagination": { "page": 1, "pageSize": 100 } }` or `{ "items": [...], "pagination": { "mode": "all", "pageSize": 100, "fetched": 250 } }`. - Validation and errors: Tool inputs are validated with Zod. Errors are formatted consistently (production messages sanitized); logs redact sensitive values. - Pagination arguments (where supported): `pageSize`, `maxPages`, `all`. The legacy `count` on `list_builds` remains for compatibility but `pageSize` is preferred. - Tools with pagination: `list_builds`, `list_projects`, `list_build_configs`, `list_vcs_roots`, `list_agents`, `list_agent_pools`, `list_test_failures`. - Enhanced results: - `get_build_results` supports options: `includeArtifacts`, `includeStatistics`, `includeChanges`, `includeDependencies`, `artifactFilter`, `maxArtifactSize`, `artifactEncoding`. - `get_build_status` supports options: `includeTests`, `includeProblems`, `includeQueueTotals`, `includeQueueReason`. - Legacy helper exports from `src/teamcity/index.ts` are retained for backwards compatibility but are not updated; use the MCP tools or `TeamCityAPI` for real workflows. ## Operation Modes ### Dev Mode Safe, read-only operations suitable for production environments. Enables monitoring, analysis, and controlled build triggering without infrastructure changes. ### Full Mode Complete infrastructure management including creation, modification, and deletion of configurations, projects, and settings. ## Tools Quick Reference | Tool Name | Description | Dev | Full | | ---------------------------- | ---------------------------------------------- | :-: | :--: | | **Build Management** | | | | | `trigger_build` | Trigger a build | ✅ | ✅ | | `get_build_status` | Get detailed build status and progress | ✅ | ✅ | | `list_builds` | Search and list builds with filtering | ✅ | ✅ | | `get_build_results` | Get detailed build results | ✅ | ✅ | | `download_build_artifact` | Download artifact content (base64/text/stream) | ✅ | ✅ | | `download_build_artifacts` | Download multiple artifacts (base64/text/stream) | ✅ | ✅ | | `fetch_build_log` | Retrieve build logs | ✅ | ✅ | | `get_build_config` | Get build configuration details | ✅ | ✅ | | `list_build_configs` | List build configurations in project | ✅ | ✅ | | **Test Analysis** | | | | | `list_test_failures` | Analyze failed tests across builds | ✅ | ✅ | | `get_test_details` | Deep dive into test results | ✅ | ✅ | | `analyze_build_problems` | Report build problems and test failures | ✅ | ✅ | | **Configuration Management** | | | | | `create_build_config` | Create new build configurations | ❌ | ✅ | | `clone_build_config` | Duplicate existing configurations | ❌ | ✅ | | `update_build_config` | Modify existing configurations | ❌ | ✅ | | `manage_build_steps` | Add, edit, remove, reorder build steps | ❌ | ✅ | | `manage_build_triggers` | Configure build triggers | ❌ | ✅ | | **VCS Management** | | | | | `list_vcs_roots` | List version control roots | ✅ | ✅ | | `create_vcs_root` | Create VCS root configurations | ❌ | ✅ | | **Project Management** | | | | | `list_projects` | List all projects | ✅ | ✅ | | `list_project_hierarchy` | Visualize project structure | ✅ | ✅ | | `create_project` | Create new projects | ❌ | ✅ | | `delete_project` | Remove projects safely | ❌ | ✅ | | `update_project_settings` | Modify project settings | ❌ | ✅ | | **Parameter Management** | | | | | `list_parameters` | List configuration parameters | ✅ | ✅ | | `add_parameter` | Add new parameters | ❌ | ✅ | | `update_parameter` | Modify existing parameters | ❌ | ✅ | | `delete_parameter` | Remove parameters | ❌ | ✅ | | **Agent Management** | | | | | `list_agents` | List build agents and status | ✅ | ✅ | | `list_agent_pools` | List agent pools | ✅ | ✅ | | `assign_agent_to_pool` | Move agents between pools | ❌ | ✅ | | `authorize_agent` | Manage agent authorization | ❌ | ✅ | | **Branch Management** | | | | | `list_branches` | List and analyze branches | ✅ | ✅ | | **Utility** | | | | | `ping` | Health check and connection test | ✅ | ✅ | **Legend:** - ✅ Available in this mode - ❌ Not available in this mode ## Tool Categories ### 1. Build Management Tools #### `trigger_build` **Description**: Trigger a TeamCity build **Mode**: Dev/Full **Key Capabilities**: - Trigger a build by build configuration ID - Optional branch selection - Optional build comment **Parameters**: - `buildTypeId` (required): Build configuration ID - `branchName` (optional): Branch to build - `comment` (optional): Build comment/description **Example Use Cases**: ```json { "buildTypeId": "MyApp_Build", "branchName": "feature/new-ui", "comment": "Testing UI changes" } ``` #### `get_build_status` **Description**: Get build status and progress **Mode**: Dev **Key Capabilities**: - Progress tracking - Stage-by-stage breakdown - Test and artifact information - Time estimations **Parameters**: - `buildId` (required): Build identifier #### `list_builds` **Description**: Search and list builds with filtering **Mode**: Dev **Key Capabilities**: - Filter by project, configuration, branch - Status and date range filtering - Pagination support - Custom field selection **Parameters**: - `projectId`: Filter by project - `buildTypeId`: Filter by configuration - `branch`: Filter by branch - `status`: SUCCESS, FAILURE, ERROR - `state`: queued, running, finished - `limit`: Maximum results - `fields`: Custom field selection #### `get_build_results` **Description**: Get comprehensive build results **Mode**: Dev **Key Capabilities**: - Test results summary - Artifact listings - Problem occurrences - Statistics and metrics **Parameters**: - `buildId` (required): Build identifier - `includeArtifacts`: Include artifact metadata alongside the build summary - `artifactEncoding`: `'base64'` (default) embeds small artifacts up to `maxArtifactSize`; `'stream'` returns a `download_build_artifact` handle instead of inline bytes - `artifactFilter`: Glob-style filter applied to artifact names - `maxArtifactSize`: Maximum size (bytes) to inline when `artifactEncoding` is `'base64'` - `includeStatistics`: Include build statistic values - `includeChanges`: Include associated VCS changes - `includeDependencies`: Include snapshot dependency builds When `artifactEncoding` is set to `'stream'`, artifacts still list metadata (`name`, `path`, `size`, `downloadUrl`) but also add a `downloadHandle` payload you can pass directly to `download_build_artifact` (or collect the `path` values for `download_build_artifacts`) to retrieve the files without embedding base64 in the response. #### `download_build_artifact` **Description**: Download a single artifact **Mode**: Dev **Key Capabilities**: - Return base64-encoded or plain-text content for small artifacts - Stream large artifacts directly to disk without buffering in memory - Enforce an optional maximum size before initiating the transfer **Parameters**: - `buildId` (required): Build identifier - `artifactPath` (required): Artifact path or name (as returned by artifact listings) - `encoding`: `'base64'` (default), `'text'`, or `'stream'` - `maxSize`: Abort download if artifact size exceeds this byte limit - `outputPath`: Absolute path to write streamed artifacts (defaults to a temp file when omitted) **Usage Notes**: - In `stream` mode the tool writes the artifact to disk and returns metadata (`outputPath`, `bytesWritten`). - Base64/text modes embed content directly in the JSON response for easy scripting. #### `download_build_artifacts` **Description**: Download multiple artifacts in a single request **Mode**: Dev **Key Capabilities**: - Batch artifact downloads to avoid repeating authentication and path resolution - Supports `'base64'`, `'text'`, and `'stream'` encodings per artifact - Shares the same streaming writer as `download_build_artifact` for consistent metadata and temp-file handling **Parameters**: - `buildId` (required): Build identifier - `artifactPaths` (required): Array of artifact paths or names to download - `encoding`: `'base64'` (default), `'text'`, or `'stream'` - `maxSize`: Abort individual downloads whose size exceeds this byte limit - `outputDir`: Absolute directory for streamed artifacts (defaults to unique temp files) **Usage Notes**: - The response contains an `artifacts` array with per-item success, encoding, and either `content` (for buffered modes) or `outputPath`/`bytesWritten` (for streaming). - Streaming mode preserves relative artifact paths under `outputDir` and will generate unique filenames when collisions occur. - Failures for individual artifacts are reported inline with `success: false` and an `error` message so callers can retry or surface partial results. #### `fetch_build_log` **Description**: Retrieve build logs **Mode**: Dev **Key Capabilities**: - Paginated retrieval by line range or explicit `startLine`/`lineCount` - Optional streaming mode that writes the response directly to disk **Parameters**: - `buildId` **or** `buildNumber` (+ optional `buildTypeId`): Identify the build to read - `page` / `pageSize`: Paged access when you do not supply `startLine` - `startLine` / `lineCount`: Explicit range when you know the offsets you need - `tail`: Return the last N lines (buffered mode only) - `encoding`: `'text'` (default) to buffer in-memory, `'stream'` to pipe to disk - `outputPath`: Destination path when streaming (defaults to a temp file) **Usage Notes**: - Streaming mode mirrors `download_build_artifact`: the tool returns metadata (`outputPath`, `bytesWritten`) instead of the log lines. - Tail mode is only available in buffered (`'text'`) encoding. ### 2. Test Analysis Tools #### `list_test_failures` **Description**: List failed tests for a build **Mode**: Dev **Key Capabilities**: - Basic listing with pagination **Parameters**: - `buildId`: Specific build or latest - `projectId`: Filter by project - `includeStackTraces`: Include full traces - `groupByError`: Group similar failures #### `get_test_details` **Description**: Get detailed information about test failures **Mode**: Dev **Key Capabilities**: - Test occurrence details for a build **Parameters**: - `testNameId` (required): Test identifier - `buildId`: Specific build context - `includeHistory`: Historical data - `limit`: History depth #### `analyze_build_problems` **Description**: Build problem and failure report **Mode**: Dev **Key Capabilities**: - Summary of problems and failed tests **Parameters**: - `buildId` (required): Build identifier - `includeInvestigations`: Include assigned investigations - `includeMuted`: Include muted problems ### 3. Configuration Management Tools (Full Mode Only) #### `create_build_config` **Description**: Create new build configurations **Mode**: Full **Key Capabilities**: - VCS root creation and attachment - Build step configuration - Trigger setup - Template-based creation - Parameter management **Parameters**: - `projectId` (required): Target project - `name` (required): Configuration name - `description`: Purpose description - `vcs`: Version control settings - `url`: Repository URL - `branch`: Default branch - `type`: git, svn, perforce - `authentication`: Credentials - `steps`: Build step array - `triggers`: Trigger array - `parameters`: Key-value parameters - `templateId`: Base template **Example**: ```typescript { projectId: "MyProject", name: "Node.js Application", vcs: { url: "https://github.com/myorg/myapp.git", branch: "main", type: "git" }, steps: [ { type: "npm", name: "Install Dependencies", script: "npm ci" }, { type: "npm", name: "Run Tests", script: "npm test" } ], triggers: [ { type: "vcs", rules: "+:*" } ] } ``` #### `clone_build_config` **Description**: Duplicate existing configurations **Mode**: Full **Key Capabilities**: - Deep or shallow cloning - Selective component copying - Name and ID customization - Cross-project cloning **Parameters**: - `sourceConfigId` (required): Source configuration - `targetProjectId` (required): Target project - `newName` (required): New configuration name - `copySettings`: Include settings - `copySteps`: Include build steps - `copyTriggers`: Include triggers - `copyDependencies`: Include dependencies #### `update_build_config` **Description**: Modify existing configurations **Mode**: Full **Key Capabilities**: - Partial updates - Setting modifications - Enable/disable configurations - Description updates **Parameters**: - `configId` (required): Configuration ID - `name`: New name - `description`: New description - `enabled`: Enable/disable - `settings`: Updated settings #### `manage_build_steps` **Description**: Comprehensive build step management **Mode**: Full **Key Capabilities**: - Add, edit, remove, reorder steps - Support for all runner types - Property configuration - Step templates **Parameters**: - `action` (required): list, add, edit, remove, reorder - `configId` (required): Configuration ID - `stepId`: Step ID (for edit/remove) - `stepConfig`: Step configuration - `name`: Step name - `type`: Runner type - `enabled`: Enable status - `properties`: Runner properties - `parameters`: Step parameters - `newOrder`: Step ID array (for reorder) **Supported Runner Types**: - `simpleRunner`: Shell/Batch scripts - `Maven2`: Maven builds - `gradle-runner`: Gradle builds - `MSBuild`: MSBuild projects - `dotnet`: .NET Core/5+ builds - `nodejs-runner`: Node.js scripts - `Docker`: Docker operations - `python`: Python scripts - `cargo`: Rust/Cargo builds - `kotlinScript`: Kotlin scripts #### `manage_build_triggers` **Description**: Configure build triggers **Mode**: Full **Key Capabilities**: - VCS triggers with branch filters - Schedule triggers (cron) - Build finish triggers - Maven snapshot dependencies **Parameters**: - `action` (required): list, add, edit, remove - `configId` (required): Configuration ID - `triggerId`: Trigger ID (for edit/remove) - `triggerConfig`: Trigger configuration - `type`: vcs, schedule, finish-build - `rules`: Branch filter rules - `schedule`: Cron expression - `buildType`: Dependent build - `branchFilter`: Branch patterns ### 4. VCS Management Tools #### `list_vcs_roots` **Description**: List version control roots **Mode**: Dev **Key Capabilities**: - Project filtering; basic listing **Parameters**: - `projectId`: Filter by project - `includeUsages`: Show where used #### `create_vcs_root` **Description**: Create VCS root configurations **Mode**: Full **Key Capabilities**: - Multi-VCS support (Git, SVN, Perforce) - Authentication configuration - Branch specification - Polling settings **Parameters**: - `projectId` (required): Target project - `name` (required): VCS root name - `url` (required): Repository URL - `type`: VCS type - `branch`: Default branch - `authentication`: Auth settings - `pollingInterval`: Check frequency ### 5. Project Management Tools #### `list_projects` **Description**: List all projects **Mode**: Dev **Key Capabilities**: - Hierarchical view - Permission filtering - Archive inclusion **Parameters**: - `includeArchived`: Include archived - `fields`: Custom fields #### `list_project_hierarchy` **Description**: Visualize project structure **Mode**: Dev/Full **Key Capabilities**: - Tree visualization - Depth control - Build configuration counts - Permission indicators **Parameters**: - `rootProjectId`: Starting point - `maxDepth`: Tree depth - `includeConfigs`: Show configurations - `format`: tree, flat, json #### `create_project` **Description**: Create new projects **Mode**: Full **Key Capabilities**: - Sub-project support - Initial configuration - Permission setup - Template application **Parameters**: - `parentProjectId` (required): Parent project - `name` (required): Project name - `id`: Custom ID - `description`: Project description - `copySettingsFrom`: Template project #### `delete_project` **Description**: Remove projects safely **Mode**: Full **Key Capabilities**: - Cascade deletion - Archive option - Safety checks - Rollback support **Parameters**: - `projectId` (required): Project to delete - `archive`: Archive instead of delete - `force`: Skip confirmations #### `update_project_settings` **Description**: Modify project settings **Mode**: Full **Key Capabilities**: - Parameter management - Feature toggling - Description updates **Parameters**: - `projectId` (required): Target project - `name`: New name - `description`: New description - `parameters`: Key-value settings - `archived`: Archive status ### 6. Parameter Management Tools #### `list_parameters` **Description**: List configuration parameters **Mode**: Dev **Key Capabilities**: - Inherited parameter tracking - Value resolution - Type information **Parameters**: - `configId` (required): Configuration ID - `includeInherited`: Show inherited - `includeSystem`: Show system params #### `add_parameter` **Description**: Add new parameters **Mode**: Full **Key Capabilities**: - Type specification - Value validation - Secure storage **Parameters**: - `configId` (required): Configuration ID - `name` (required): Parameter name - `value` (required): Parameter value - `type`: text, password, select - `label`: Display label - `description`: Help text #### `update_parameter` **Description**: Modify existing parameters **Mode**: Full **Key Capabilities**: - Value updates - Type changes - Metadata updates **Parameters**: - `configId` (required): Configuration ID - `name` (required): Parameter name - `value`: New value - `type`: New type - `description`: New description #### `delete_parameter` **Description**: Remove parameters **Mode**: Full **Key Capabilities**: - Safe deletion - Dependency checking **Parameters**: - `configId` (required): Configuration ID - `name` (required): Parameter name ### 7. Agent Management Tools #### `list_agents` **Description**: List build agents **Mode**: Dev **Key Capabilities**: - Status monitoring - Capability listing - Pool assignments - Running builds **Parameters**: - `includeDisconnected`: Include offline - `poolId`: Filter by pool - `authorized`: Filter by auth status #### `list_agent_pools` **Description**: List agent pools **Mode**: Dev/Full **Key Capabilities**: - Basic listing **Parameters**: - `includeProjects`: Show projects - `includeAgents`: List agents #### `assign_agent_to_pool` **Description**: Move agents between pools **Mode**: Full **Key Capabilities**: - Pool reassignment - Bulk operations **Parameters**: - `agentId` (required): Agent ID - `poolId` (required): Target pool #### `authorize_agent` **Description**: Manage agent authorization **Mode**: Full **Key Capabilities**: - Authorize/unauthorize - Comment addition **Parameters**: - `agentId` (required): Agent ID - `authorized` (required): Auth status - `comment`: Reason/comment ### 8. Branch Management Tools #### `list_branches` **Description**: List branches from recent builds **Mode**: Dev **Key Capabilities**: - Branch names discovered from recent builds **Parameters**: - `configId` (required): Configuration ID - `mode`: all, active, default, with-builds - `includeBuilds`: Include build info ### 9. Utility Tools #### `ping` **Description**: Connectivity test **Mode**: Dev **Key Capabilities**: - Server connectivity echo **Parameters**: Optional `message` ## Common Workflows ### 1. Continuous Integration Workflow ```yaml Workflow: Set up CI for new project Steps: 1. Create project structure - create_project (parent, name, description) 2. Create VCS root - create_vcs_root (git URL, authentication) 3. Create build configuration - create_build_config (project, VCS, steps) 4. Configure triggers - manage_build_triggers (add VCS trigger) 5. Test the setup - trigger_build (specify branch/comment as needed) 6. Monitor first build - get_build_status ``` ### 2. Test Failure Investigation ```yaml Workflow: Investigate test failures Steps: 1. List recent failed builds - list_builds (status: FAILURE) 2. Get test failures - list_test_failures (buildId, includeStackTraces) 3. Analyze specific test - get_test_details (testNameId, includeHistory) 4. Review build problems - analyze_build_problems (buildId) 5. Check build logs - fetch_build_log (buildId, errors-only) ``` ### 3. Deployment Pipeline Setup ```yaml Workflow: Create deployment pipeline Steps: 1. Create staging configuration - clone_build_config (from dev config) 2. Add deployment steps - manage_build_steps (add Docker/K8s steps) 3. Configure parameters - add/update_parameter (environment vars) 4. Set up triggers - manage_build_triggers (finish-build from tests) 5. Create production config - clone_build_config (from staging) 6. Add manual trigger requirement - update_build_config (require approval) ``` ### 4. Build Performance Optimization ```yaml Workflow: Optimize slow builds Steps: 1. Analyze build history - list_builds (with timing data) 2. Review build steps - manage_build_steps (list with timings) 3. Check agent utilization - list_agents (with current builds) 4. Examine test performance - get_test_details (performance analysis) 5. Reconfigure steps - manage_build_steps (reorder/parallelize) 6. Update agent pools - assign_agent_to_pool (balance load) ``` ### 5. Multi-Branch Development ```yaml Workflow: Support feature branches Steps: 1. List active branches - list_branches (mode: active) 2. Configure branch filters - manage_build_triggers (branch patterns) 3. Set up personal builds - trigger_build (personal: true) 4. Monitor branch builds - list_builds (branch filter) 5. Clean up old branches - list_branches (identify inactive) ``` ## Error Handling ### Common Error Patterns 1. **Authentication Errors** - Invalid token: Check TEAMCITY_TOKEN - Expired token: Regenerate in TeamCity - Permission denied: Verify user roles 2. **Configuration Errors** - Not found: Verify ID/name - Already exists: Use unique names - Invalid parameters: Check schema 3. **Network Errors** - Connection refused: Verify TEAMCITY_URL - Timeout: Check network/firewall - SSL errors: Verify certificates 4. **Validation Errors** - Schema validation: Check input format - Missing required: Provide all required fields - Type mismatch: Verify data types ### Error Recovery Strategies 1. **Automatic Retry**: Network failures retry with exponential backoff 2. **Caching**: Failed requests served from cache when available 3. **Fallback**: Degraded functionality in dev mode 4. **Rollback**: Configuration changes can be reverted ## Security Considerations ### Authentication - Token-based authentication only - No password storage - Secure token transmission ### Authorization - Mode-based permission control - Project-level access control - Audit logging for all changes ### Data Protection - Sensitive parameter encryption - Secure VCS credential storage - No credential logging ## Testing Strategies ### Unit Testing ```typescript // Test individual tool handlers describe('trigger_build', () => { it('should queue build successfully', async () => { const result = await triggerBuildTool.handler({ buildConfiguration: 'Test_Build', branch: 'main', }); expect(result.success).toBe(true); }); }); ``` ### Integration Testing ```typescript // Test complete workflows describe('CI Workflow', () => { it('should set up complete CI pipeline', async () => { // Create project const project = await createProject({ ... }); // Add configuration const config = await createBuildConfig({ ... }); // Trigger build const build = await triggerBuild({ ... }); // Verify results expect(build.status).toBe('SUCCESS'); }); }); ``` ### End-to-End Testing ```typescript // Test through MCP protocol describe('MCP Integration', () => { it('should handle tool calls via MCP', async () => { const response = await mcpClient.callTool('trigger_build', { buildConfiguration: 'E2E_Test' }); expect(response.success).toBe(true); }); }); ``` ## Best Practices ### 1. Configuration as Code - Store configurations in version control - Use templates for consistency - Automate configuration updates ### 2. Build Organization - Group related builds in projects - Use clear naming conventions - Document build purposes ### 3. Performance - Parallelize independent steps - Use agent pools effectively - Cache dependencies ### 4. Monitoring - Set up build failure notifications - Track build time trends - Monitor test stability ### 5. Security - Rotate tokens regularly - Use parameter encryption - Audit configuration changes ## Troubleshooting Guide ### Build Failures 1. Check `fetch_build_log` for errors 2. Review `analyze_build_problems` 3. Examine `list_test_failures` 4. Verify configuration with `get_build_config` ### Performance Issues 1. Analyze with `get_build_results` 2. Check agent availability 3. Review step timings 4. Optimize parallelization ### Configuration Problems 1. Validate with dry run 2. Check parameter values 3. Verify VCS access 4. Review trigger rules ## Appendix ### A. Runner Type Properties #### Shell/Batch Script (simpleRunner) - `script.content`: Script body - `use.custom.script`: true/false - `teamcity.step.mode`: default/execute #### Maven (Maven2) - `goals`: Maven goals - `pomLocation`: POM file path - `maven.home`: Maven installation #### Gradle (gradle-runner) - `ui.gradleRunner.gradle.tasks.names`: Tasks - `ui.gradleRunner.gradle.wrapper.path`: Wrapper path - `ui.gradleRunner.gradle.wrapper.useWrapper`: true/false #### Docker - `docker.image.platform`: Platform - `docker.command.type`: build/run/push - `dockerfile.path`: Dockerfile location ### B. Trigger Rule Syntax #### VCS Trigger Rules - `+:*` - All branches - `+:refs/heads/main` - Main branch only - `+:refs/heads/feature/*` - Feature branches - `-:refs/heads/experimental/*` - Exclude experimental #### Schedule Expressions - `0 0 * * *` - Daily at midnight - `0 */2 * * *` - Every 2 hours - `0 9-17 * * 1-5` - Weekdays 9-5 ### C. Parameter Types - `text` - Plain text - `password` - Encrypted storage - `checkbox` - Boolean - `select` - Dropdown with options - `display` - Read-only display ### D. API Response Formats All tools return a consistent response format: ```typescript interface ToolResponse { success: boolean; data?: any; error?: string; content?: string; } ``` ## Conclusion The TeamCity MCP server provides CI/CD automation via MCP tools. With proper understanding of these tools and workflows, teams can achieve efficient, reliable continuous integration and deployment processes. For updates and additional information, refer to the project repository and TeamCity documentation.