@bldbl/mcp

Implementation Reference

• src/cli.ts:162-213 (registration)

  MCP tool registration for 'complete_task' using McpServer.tool(), including inline Zod input schema and async handler function.
  this.server.tool( 'complete_task', { task_id: z.string().describe('The ID of the task to complete'), completion_notes: z.string().describe('Notes about task completion'), files_modified: z .array(z.string()) .optional() .describe('List of files that were modified'), testing_completed: z .boolean() .optional() .describe('Whether testing was completed'), documentation_updated: z .boolean() .optional() .describe('Whether documentation was updated'), time_spent: z .number() .optional() .describe('Total time spent in minutes'), }, async ({ task_id, completion_notes, files_modified, testing_completed, documentation_updated, time_spent, }) => { if (!this.client) { throw new Error('Not connected to Buildable API'); } const result = await this.client.completeTask(task_id, { completion_notes, files_modified: files_modified || [], testing_completed: testing_completed || false, documentation_updated: documentation_updated || false, time_spent: time_spent || 0, }); return { content: [ { type: 'text', text: JSON.stringify(result, null, 2), }, ], }; } );
• src/cli.ts:184-212 (handler)

  Handler function that executes the complete_task tool logic: checks client connection, calls BuildableMCPClient.completeTask API method, formats result as MCP content response.
  async ({ task_id, completion_notes, files_modified, testing_completed, documentation_updated, time_spent, }) => { if (!this.client) { throw new Error('Not connected to Buildable API'); } const result = await this.client.completeTask(task_id, { completion_notes, files_modified: files_modified || [], testing_completed: testing_completed || false, documentation_updated: documentation_updated || false, time_spent: time_spent || 0, }); return { content: [ { type: 'text', text: JSON.stringify(result, null, 2), }, ], }; }
• src/cli.ts:164-183 (schema)

  Zod schema defining the input parameters for the complete_task MCP tool.
  { task_id: z.string().describe('The ID of the task to complete'), completion_notes: z.string().describe('Notes about task completion'), files_modified: z .array(z.string()) .optional() .describe('List of files that were modified'), testing_completed: z .boolean() .optional() .describe('Whether testing was completed'), documentation_updated: z .boolean() .optional() .describe('Whether documentation was updated'), time_spent: z .number() .optional() .describe('Total time spent in minutes'), },
• src/client.ts:195-226 (helper)

  Supporting client method BuildableMCPClient.completeTask that performs the actual HTTP POST request to the Buildable API to complete the task.
  async completeTask( taskId: string, completion: CompleteTaskRequest ): Promise<CompleteTaskResponse> { this.log('debug', `Completing task ${taskId}...`); try { const response = await this.makeRequest<CompleteTaskResponse>( 'POST', `/tasks/${taskId}/complete`, { files_created: completion.files_modified, files_modified: completion.files_modified, completion_notes: completion.completion_notes, time_spent_minutes: completion.time_spent, verification_evidence: completion.testing_completed ? 'Tests passed' : undefined, } ); this.log('info', `Successfully completed task ${taskId}`); // Update connection status back to 'connected' await this.updateConnectionStatus('connected'); return response.data!; } catch (error) { this.log('error', `Failed to complete task ${taskId}:`, error); throw error; } }
• src/types.ts:116-125 (schema)

  TypeScript interface CompleteTaskRequest defining the structure for task completion data used by the client method.
  export interface CompleteTaskRequest { completion_notes: string; files_modified: string[]; testing_completed: boolean; documentation_updated: boolean; time_spent: number; // minutes challenges_faced?: string[]; lessons_learned?: string[]; next_recommendations?: string[]; }