claude-code-sdk-ts

🥈Silver

Fluent, chainable TypeScript SDK: configure models, enable tools, stream events, then fetch text, JSON, run details or token stats in one call via .asText() or .allowTools('Read', 'Write'). Multi-level logging plus live onMessage/onToolUse callbacks give deep, CLI-compatible observability.

207270Updated 2w ago

Intermediate30min to implementautomation

Saves ~90 min per use

Quick InstallView Source

git clone https://github.com/instantlyeasy/claude-code-sdk-ts.git

Works with:

Claude

Overview

About This Skill

How to Use

[{"step":"Install the SDK and set up your project","action":"Run `npm install @anthropic-ai/claude-code-sdk-ts` in your project directory. Ensure you have a valid Anthropic API key configured in your environment variables as `ANTHROPIC_API_KEY`.","tip":"Use `tsc --init` to generate a TypeScript config if starting a new project."},{"step":"Configure the SDK for your task","action":"Set model parameters (e.g., `model`, `maxTokens`, `temperature`) and enable required tools (e.g., `Read`, `Write`, `Bash`). Chain these configurations before calling `.run()`.","tip":"Enable only the tools you need to reduce costs and improve performance. For example, omit `Bash` if you don't need shell access."},{"step":"Implement callbacks for observability","action":"Add `.onMessage()` to log streaming text and `.onToolUse()` to track tool invocations. These callbacks provide real-time feedback during execution.","tip":"Use these callbacks to update a progress bar in your CLI or write logs to a file for debugging."},{"step":"Execute your task and fetch results","action":"Call `.run()` with an async function that uses the enabled tools. Retrieve the final output with `.asText()` for text responses or `.allowTools('Read', 'Write')` to get structured tool outputs.","tip":"For complex workflows, chain multiple `.run()` calls or use the returned `tools` object to perform sequential operations."},{"step":"Analyze the results","action":"Inspect the returned object for `output`, `stats` (token usage), and `run` (execution details). Use this data to validate success or debug issues.","tip":"Log token stats to monitor API usage and costs. Compare `inputTokens` and `outputTokens` against your budget."}]

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/instantlyeasy/claude-code-sdk-ts

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use the claude-code-sdk-ts to automate [TASK] in [PROJECT_PATH]. Configure the model with [MODEL_SETTINGS] and enable tools: [TOOL_LIST]. Stream events to [LOGGING_TARGET]. Fetch the final output as [OUTPUT_FORMAT] using .asText() or .allowTools('Read', 'Write'). Include token stats and run details in the response.

Example Output

```typescript
import { ClaudeCode } from '@anthropic-ai/claude-code-sdk-ts';

const result = await new ClaudeCode()
  .configure({
    model: 'claude-3-7-sonnet-20250219',
    maxTokens: 4096,
    temperature: 0.3,
  })
  .enableTools('Read', 'Write', 'Bash')
  .onMessage((msg) => console.log('Streaming:', msg.text))
  .onToolUse((tool) => console.log('Tool used:', tool.name))
  .run(async ({ tools }) => {
    await tools.Write('src/utils/helper.ts', 'export const greet = () => "Hello";');
    const content = await tools.Read('src/utils/helper.ts');
    return content;
  })
  .asText();

console.log('Output:', result.output);
console.log('Token stats:', result.stats);
console.log('Run details:', result.run);
```

**Output:**
```
Output: export const greet = () => "Hello";
Token stats: {
  inputTokens: 128,
  outputTokens: 45,
  totalTokens: 173,
  cacheCreationInputTokens: 0,
  cacheReadInputTokens: 0
}
Run details: {
  durationMs: 1420,
  model: 'claude-3-7-sonnet-20250219',
  toolsUsed: ['Write', 'Read'],
  steps: 3
}
```

**Logging Output:**
```
Streaming: Starting file write operation...
Tool used: Write
Streaming: File written successfully.
Streaming: Reading file content...
Tool used: Read
Streaming: File content retrieved.
```