
from ash-ai32
SDK skill for deploying and managing hosted AI agents with Ash: create clients and sessions, stream SSE responses, handle lifecycle and errors, using TypeScript
The Ash SDK skill documents how to deploy and orchestrate hosted AI agents using the Ash platform. It covers creating an Ash client, deploying agents, creating and managing sessions, sending messages, and consuming real-time SSE streaming responses (including partial text deltas). The skill provides practical code examples in TypeScript and Python and a validation checklist for production usage.
Use this skill when you need a reproducible integration with Ash: programmatic agent deployment, session lifecycle management (create, pause, resume, end), streaming agent output for UIs or real-time pipelines, or handling SDK-level errors and workspace files. It is appropriate for backend services, developer tooling, and server integrations.
references/ directory is included with deeper docs on streaming, sessions, and API surface.Compatible with agent orchestration and developer agents that invoke SDKs (TypeScript, Python) and systems expecting SSE streams.
Ash SDK skill provides clear documentation for deploying and managing hosted AI agents via the Ash platform, with TypeScript and Python SDK examples. No bundled scripts to test — purely a reference/documentation skill. SKILL.md is well-structured with a decision workflow, quick start guide, streaming examples, and pointers to detailed references. Clean security posture with no suspicious patterns.
Documentation-only skill with no executable scripts. Code examples use placeholder credentials. Well-organized with progressive disclosure via references/ directory.