skills.rest
tangle-network
tangle-networkOFFICIAL
Explore all skills by this author →
Website𝕏@tangle_networkGitHub

sandbox-sdk

Manage agent sessions and configure provider backends for sandboxed AI orchestration.

1|Updated Mar 7, 2026
Search Tags:#telemetry#agent-sessions#workspace-management#provider-adapters#sandbox-sdk#sse-streaming#auth-materialization
One-click install
npx skills add https://github.com/tangle-network/skills --skill sandbox-sdk-tangle-network
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sandbox-sdk
Source: https://github.com/tangle-network/skills/tree/main/plugins/sandbox-sdk/skills/sandbox-sdk
Command: npx skills add https://github.com/tangle-network/skills --skill sandbox-sdk-tangle-network

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Building and operating agent-driven applications on the Tangle Sandbox SDK involves many moving parts: configuring provider backends, managing sessions, streaming event data, handling CLI OAuth materialization, and collecting telemetry. This Skill consolidates best practices, primitives, and operational patterns to reduce integration friction and runtime errors when developing against the sandbox stack.

Core Features & Use Cases

  • Backend configuration & connection: Guidance for server-based and CLI-based provider integrations, profile usage, and cache keys.
  • Session lifecycle management: Create, fork, execute, stream, persist, and shutdown agent sessions with workspace and auth resolution.
  • SSE streaming & parsing: Use provided SSEChunkParser and parsing primitives to reliably stream and convert events into trace and signal formats.
  • Auth materialization & security: Materialize CLI OAuth/auth files into sandbox HOME with safe path validation and permission handling.
  • Telemetry & observability: Integrate trace sinks, signals, and usage tracking for Langfuse, OTEL, HTTP, or console sinks.
  • Provider adapter development: Implement SdkProviderAdapter or extend the CLI base class to add new provider adapters and MCP integrations.

Quick Start

Use the sandbox-sdk to create a new agent session that connects to an OpenCode backend, stream SSE output to the client, and collect telemetry for the execution.

Frequently Asked Questions about sandbox-sdk

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I manage agent sessions for sandboxed AI orchestration?▼

Managing agent sessions for sandboxed AI orchestration involves creating, forking, executing, streaming, persisting, and shutting down sessions. This requires proper workspace and auth resolution to ensure robust production deployments and reliable session lifecycles.

What is the best way to parse SSE streaming events for AI agents?▼

Parsing SSE streaming events for AI agents is best handled using an SSEChunkParser. This primitive reliably streams and converts server-sent events into trace and signal formats, ensuring accurate telemetry and observability data collection during execution.

How do I materialize CLI auth files securely in a sandbox environment?▼

Materializing CLI auth files in a sandbox environment safely requires writing OAuth materials into the sandbox HOME directory. This process demands strict path validation and proper permission handling to maintain security and prevent unauthorized file access.

How can I integrate telemetry and observability into my provider backends?▼

Integrating telemetry and observability into provider backends requires applying trace sinks and signals for usage tracking. You can route this telemetry data to Langfuse, OTEL, HTTP, or console sinks to monitor execution and maintain system health.

Does this SDK support adding custom provider adapters for new integrations?▼

Yes, adding custom provider adapters for new integrations is supported. You can implement the SdkProviderAdapter interface or extend the CLI base class to build new provider adapters and configure MCP integrations for your specific backend requirements.

What are the limitations when configuring backend connections for agent sessions?▼

Configuring backend connections for agent sessions has limitations regarding cache keys and profile usage. Both server-based and CLI-based provider integrations require careful workspace resolution and backend caching configuration to avoid runtime errors during execution.

Related Skills

Explore More Software Engineering Skills→
johnnystefanjohnnystefan

prisma-postgres

Provision Prisma Postgres databases via console, create-db CLI, or management API workflows.

Community
Intermediate
sammcjsammcj
155

aws-strands-agents-agentcore

Guide AWS Strands Agents SDK and AgentCore deployments across Lambda, ECS/Fargate, and AgentCore Runtime.

Community
Advanced
Ikram-AlamIkram-Alam

streaming-llm-responses

Stream LLM responses with lifecycle events and thread-state synchronization.

Community
Advanced
rblake2320rblake2320

nemoclaw-user-deploy-remote

Deploy NemoClaw to a remote GPU instance and onboard the sandbox.

Community
Intermediate