agentsclimarketplace

Twinmind local dev loop

Skill jeremylongshore/claude-code-plugins-plus-skills/plugins/saas-packs/twinmind-pack/skills/twinmind-local-dev-loop

'Set up local development workflow with TwinMind API integration.From its SKILL.md

Install
npx -y skills add jeremylongshore/claude-code-plugins-plus-skills --skill twinmind-local-dev-loop

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

8.2 KB, ~1.9k tokens by cl100k_base, as published. Nobody here has run it

TwinMind Local Dev Loop

Overview

Configure a productive local development environment for TwinMind API integration.

Prerequisites

  • TwinMind Pro or Enterprise account (API access)
  • Node.js 18+ or Python 3.10+
  • API key from TwinMind dashboard
  • Local development environment

Instructions

Step 1: Project Setup

set -euo pipefail
# Create project directory
mkdir twinmind-integration && cd twinmind-integration

# Initialize Node.js project
npm init -y

# Install dependencies
npm install dotenv axios zod typescript ts-node @types/node

# Initialize TypeScript
npx tsc --init

Step 2: Configure Environment

# Create environment file
cat > .env << 'EOF'
TWINMIND_API_KEY=your-api-key-here
TWINMIND_API_URL=https://api.twinmind.com/v1
TWINMIND_WEBHOOK_SECRET=your-webhook-secret
NODE_ENV=development
EOF

# Add to .gitignore
echo ".env" >> .gitignore
echo "node_modules" >> .gitignore

Step 3: Create TwinMind Client

// src/twinmind/client.ts
import axios, { AxiosInstance } from 'axios';
import { z } from 'zod';

// Response schemas
const TranscriptSchema = z.object({
  id: z.string(),
  text: z.string(),
  duration_seconds: z.number(),
  language: z.string(),
  speakers: z.array(z.object({
    id: z.string(),
    name: z.string().optional(),
    segments: z.array(z.object({
      start: z.number(),
      end: z.number(),
      text: z.string(),
      confidence: z.number(),
    })),
  })),
  created_at: z.string(),
});

const SummarySchema = z.object({
  id: z.string(),
  transcript_id: z.string(),
  summary: z.string(),
  action_items: z.array(z.object({
    text: z.string(),
    assignee: z.string().optional(),
    due_date: z.string().optional(),
  })),
  key_points: z.array(z.string()),
});

export type Transcript = z.infer<typeof TranscriptSchema>;
export type Summary = z.infer<typeof SummarySchema>;

export class TwinMindClient {
  private client: AxiosInstance;

  constructor(apiKey: string, baseUrl?: string) {
    this.client = axios.create({
      baseURL: baseUrl || 'https://api.twinmind.com/v1',
      headers: {
        'Authorization': `Bearer ${apiKey}`,
        'Content-Type': 'application/json',
      },
      timeout: 30000,  # 30000: 30 seconds in ms
    });
  }

  async healthCheck(): Promise<boolean> {
    const response = await this.client.get('/health');
    return response.status === 200;  # HTTP 200 OK
  }

  async transcribe(audioUrl: string, options?: {
    language?: string;
    diarization?: boolean;
    model?: 'ear-3' | 'ear-2';
  }): Promise<Transcript> {
    const response = await this.client.post('/transcribe', {
      audio_url: audioUrl,
      language: options?.language || 'auto',
      diarization: options?.diarization ?? true,
      model: options?.model || 'ear-3',
    });
    return TranscriptSchema.parse(response.data);
  }

  async summarize(transcriptId: string): Promise<Summary> {
    const response = await this.client.post('/summarize', {
      transcript_id: transcriptId,
    });
    return SummarySchema.parse(response.data);
  }

  async search(query: string, options?: {
    limit?: number;
    date_from?: string;
    date_to?: string;
  }): Promise<Transcript[]> {
    const response = await this.client.get('/search', {
      params: {
        q: query,
        limit: options?.limit || 10,
        date_from: options?.date_from,
        date_to: options?.date_to,
      },
    });
    return z.array(TranscriptSchema).parse(response.data.results);
  }
}

Step 4: Create Dev Runner Script

// src/dev.ts
import 'dotenv/config';
import { TwinMindClient } from './twinmind/client';

async function main() {
  const client = new TwinMindClient(process.env.TWINMIND_API_KEY!);

  // Health check
  console.log('Checking TwinMind API health...');
  const healthy = await client.healthCheck();
  console.log(`API Status: ${healthy ? 'Healthy' : 'Unhealthy'}`);

  // Example: Transcribe audio file
  // const transcript = await client.transcribe('https://example.com/meeting.mp3');
  // console.log('Transcript:', transcript);

  // Example: Generate summary
  // const summary = await client.summarize(transcript.id);
  // console.log('Summary:', summary);

  // Example: Search memory vault
  // const results = await client.search('budget review');
  // console.log('Search results:', results);
}

main().catch(console.error);

Step 5: Configure Dev Scripts

// package.json scripts
{
  "scripts": {
    "dev": "ts-node src/dev.ts",
    "build": "tsc",
    "test": "jest",
    "lint": "eslint src/**/*.ts",
    "typecheck": "tsc --noEmit"
  }
}

Step 6: Create Test Fixtures

// tests/fixtures/mock-responses.ts
export const mockTranscript = {
  id: 'tr_test_123',
  text: 'Welcome to the meeting. Today we discuss the project timeline.',
  duration_seconds: 45,
  language: 'en',
  speakers: [
    {
      id: 'spk_1',
      name: 'Host',
      segments: [
        {
          start: 0,
          end: 5.2,
          text: 'Welcome to the meeting.',
          confidence: 0.97,
        },
        {
          start: 5.5,
          end: 12.1,
          text: 'Today we discuss the project timeline.',
          confidence: 0.95,
        },
      ],
    },
  ],
  created_at: '2025-01-15T10:00:00Z',  # 2025 year
};

export const mockSummary = {
  id: 'sum_test_456',
  transcript_id: 'tr_test_123',
  summary: 'Brief meeting to discuss project timeline and milestones.',
  action_items: [
    {
      text: 'Review timeline document',
      assignee: 'John',
      due_date: '2025-01-20',
    },
  ],
  key_points: [
    'Project kickoff confirmed',
    'Timeline review scheduled',
  ],
};

Step 7: Run Development Loop

set -euo pipefail
# Start development
npm run dev

# Watch mode (requires nodemon)
npm install -D nodemon
npx nodemon --exec ts-node src/dev.ts

# Run with debug logging
DEBUG=twinmind:* npm run dev

Output

  • Project structure with TypeScript configuration
  • TwinMind client with type-safe schemas
  • Environment configuration
  • Development scripts
  • Test fixtures for offline development

Error Handling

ErrorCauseSolution
API key missing.env not loadedCheck dotenv import
Connection refusedWrong API URLVerify TWINMIND_API_URL
Rate limitedToo many requestsAdd delay between calls
Schema validation failedAPI response changedUpdate Zod schemas
TimeoutLarge audio fileIncrease timeout value

Project Structure

twinmind-integration/
├── src/
│   ├── twinmind/
│   │   ├── client.ts       # API client
│   │   ├── types.ts        # TypeScript types
│   │   └── errors.ts       # Error classes
│   ├── services/
│   │   └── meeting.ts      # Business logic
│   └── dev.ts              # Development entry
├── tests/
│   ├── fixtures/
│   │   └── mock-responses.ts
│   └── twinmind.test.ts
├── .env                    # Environment (gitignored)
├── .env.example            # Template
├── package.json
└── tsconfig.json

Resources

Next Steps

Apply patterns in twinmind-sdk-patterns for production-ready code.

Examples

Basic usage: Apply twinmind local dev loop to a standard project setup with default configuration options.

Advanced scenario: Customize twinmind local dev loop for production environments with multiple constraints and team-specific requirements.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.