@DanKornas: Building AI agents gets harder when flexible LLM behavior and predictable execution have to coexist. AgentDock is a Typ…
Summary
AgentDock is an open-source TypeScript framework for building configurable AI agents, combining flexible LLM behavior with deterministic execution through node-based architecture.
View Cached Full Text
Cached at: 07/25/26, 08:12 PM
Building AI agents gets harder when flexible LLM behavior and predictable execution have to coexist.
AgentDock is a TypeScript, backend-first framework for developers building configurable AI agents and agent-powered applications.
It helps you combine adaptive LLM decisions with controlled execution by modeling capabilities as nodes and defining deterministic tool paths or sub-workflows where consistency matters.
Key features: • Node-based architecture – build agents from BaseNode, AgentNode, tools, and custom nodes. • Configurable determinism – choose where LLM inference varies and where workflows stay defined. • Provider-independent core – connect different LLM providers through a unified interface. • Orchestration and sessions – control tool availability by context and isolate conversation state. • Reference implementation – inspect a complete Next.js client plus preconfigured example agents.
It’s open-source (MIT license).
Link in the reply
Similar Articles
@DanKornas: Agent projects get hard to maintain when tools, memory, workflows, and model integrations all live in separate glue cod…
VoltAgent is an open-source TypeScript framework for building AI agents and multi-agent systems, organizing tools, memory, workflows, and model integrations in one cohesive framework.
AgentsDock: An IDE designed for agentic AI research
AgentsDock is a multi-platform IDE for agentic AI research that enables agent management, server connectivity, and integrated terminal access.
@DanKornas: Building an AI agent in Go shouldn’t mean wiring every model, tool, and memory layer from scratch. Agent Go SDK is a Go…
Agent Go SDK is an open-source Go framework for building configurable AI agents with support for multiple model providers, memory backends, tools, MCP connections, and declarative YAML configuration.
The most reliable data agent I've shipped is ~90% deterministic code. The LLM just parses intent and talks. Change my mind.
The author argues that the reliability of AI agents comes from deterministic code, not the LLM, and shares five key practices for building trustworthy agents on messy real-world data.
@DanKornas: AI workflow stacks get messy fast. This repo keeps the pieces together. Output is an open-source TypeScript framework f…
Output is an open-source TypeScript framework for building AI workflows and agents aimed at Claude Code users, keeping prompts, evaluations, tracing, cost tracking, orchestration, and credentials unified in one codebase. Key features include workflow folders, version-controlled prompts, automatic tracing, LLM-as-judge evaluations, and durable orchestration powered by Temporal, released under the Apache 2.0 license.