https://github.com/declaw-ai/declaw-js
TypeScript SDK for Declaw — security-first sandboxing for AI agents
https://github.com/declaw-ai/declaw-js
agentic-ai ai-agents ai-security code-execution code-interpreter firecracker guardrails javascript llm llm-security microvm nodejs prompt-injection sandbox sandboxing sdk typescript
Last synced: 11 days ago
JSON representation
TypeScript SDK for Declaw — security-first sandboxing for AI agents
- Host: GitHub
- URL: https://github.com/declaw-ai/declaw-js
- Owner: declaw-ai
- License: apache-2.0
- Created: 2026-06-11T12:32:13.000Z (17 days ago)
- Default Branch: main
- Last Pushed: 2026-06-16T23:04:57.000Z (12 days ago)
- Last Synced: 2026-06-17T01:02:45.332Z (12 days ago)
- Topics: agentic-ai, ai-agents, ai-security, code-execution, code-interpreter, firecracker, guardrails, javascript, llm, llm-security, microvm, nodejs, prompt-injection, sandbox, sandboxing, sdk, typescript
- Language: TypeScript
- Homepage: https://declaw.ai
- Size: 217 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# @declaw/sdk
Secure runtime for AI agents. Spin up isolated sandboxes in milliseconds with built-in guardrails — PII scanning, prompt injection defense, network isolation, and egress filtering.
## Install
```bash
npm install @declaw/sdk
```
## Quick Start
```typescript
import { Sandbox } from '@declaw/sdk';
const sandbox = await Sandbox.create({
apiKey: 'your-api-key',
template: 'base',
timeout: 60,
});
// Run commands
const result = await sandbox.commands.run('echo "Hello from a secure sandbox"');
console.log(result.stdout);
// Read/write files
await sandbox.files.write('/tmp/hello.txt', 'Hello World');
const content = await sandbox.files.read('/tmp/hello.txt');
// Clean up
await sandbox.kill();
```
## Why Declaw?
AI agents need to execute code, call APIs, and interact with the world. Declaw gives them a secure sandbox to do it — with built-in guardrails that protect your users and infrastructure.
- **Sub-10ms sandbox creation** — pre-warmed VM pool, no cold starts
- **Network isolation** — per-sandbox firewall with domain and CIDR rules
- **Full file system** — read, write, upload, download files in the sandbox
## Security & Guardrails
Every outbound request from the sandbox passes through a configurable security pipeline.
### PII Scanning
Detect and redact sensitive data before it leaves the sandbox.
```typescript
const sandbox = await Sandbox.create({
security: SecurityPolicy.from({
pii: {
enabled: true,
types: ['ssn', 'credit_card', 'email', 'phone', 'api_key'],
action: 'redact',
},
}),
});
```
### Prompt Injection Defense
Block prompt injection attempts in agent outputs.
```typescript
const sandbox = await Sandbox.create({
security: SecurityPolicy.from({
injectionDefense: {
enabled: true,
action: 'block',
threshold: 0.85,
},
}),
});
```
### Toxicity Filtering
```typescript
security: SecurityPolicy.from({
toxicity: { enabled: true, action: 'block', threshold: 0.7 },
})
```
### Code Security & Invisible Text Detection
```typescript
security: SecurityPolicy.from({
codeSecurity: { enabled: true, action: 'log' },
invisibleText: { enabled: true, action: 'block' },
})
```
### Network Policies
```typescript
// Allow only specific domains
const sandbox = await Sandbox.create({
network: { allowOut: ['api.openai.com', 'huggingface.co'] },
});
// Block all egress
const isolated = await Sandbox.create({
network: { denyOut: ['ALL_TRAFFIC'] },
});
```
### Data Transformation
Transform sensitive values in-flight.
```typescript
security: SecurityPolicy.from({
transformations: [
{ pattern: 'sk-[a-zA-Z0-9]+', replacement: '[API_KEY]', direction: 'egress' },
],
})
```
### Combining Guardrails
All guardrails compose — enable multiple and they run in sequence:
```typescript
const sandbox = await Sandbox.create({
template: 'ai-agent',
timeout: 300,
network: { allowOut: ['api.openai.com', 'api.anthropic.com'] },
security: SecurityPolicy.from({
pii: { enabled: true, action: 'redact', types: ['ssn', 'credit_card'] },
injectionDefense: { enabled: true, action: 'block' },
toxicity: { enabled: true, action: 'log' },
invisibleText: { enabled: true, action: 'block' },
}),
});
```
## Templates
| Template | Description |
|----------|-------------|
| `base` | Minimal Linux |
| `python` | Python 3.12 with pip |
| `node` | Node.js 22 LTS with npm |
| `code-interpreter` | Python with data science libraries |
| `ai-agent` | Python + Node.js + AI/ML tools |
| `mcp-server` | MCP server runtime |
| `web-dev` | Node.js + browser testing |
| `devops` | Docker, Terraform, kubectl |
## API
```typescript
// Create sandbox
const sandbox = await Sandbox.create({ template, apiKey, timeout, network, security });
// Commands
const result = await sandbox.commands.run('ls -la');
const stream = sandbox.commands.stream('python script.py');
// Files — `path` is the literal absolute path inside the sandbox.
// Files appear at exactly that path — no remapping, no bridge directory.
await sandbox.files.write(path, content);
const data = await sandbox.files.read(path);
const entries = await sandbox.files.list('/');
// PTY (interactive terminal)
const pty = await sandbox.pty.create({ cols: 80, rows: 24 });
// Lifecycle
await sandbox.kill();
```
## License
Apache-2.0