cacheplane/threadplane

The open-source thread-plane for agents: chat, durable threads, persistence, human approvals, and generative UI for Angular, on LangGraph and AG-UI.

What it solves

Threadplane provides a standardized way to build AI agent chat interfaces in Angular (versions 20-22). It solves the problem of connecting complex agent backends—which often involve streaming, human-in-the-loop approvals, and dynamic UI elements—to a frontend without writing repetitive boilerplate or locking the UI into a specific backend implementation.

How it works

Threadplane introduces a runtime-neutral Agent contract that acts as a bridge between the UI and the backend. It uses adapters (such as for LangGraph or AG-UI) to translate backend-specific streams into Angular Signals. This allows developers to bind agent state (like messages, loading status, and tool progress) directly to the UI reactively. It also includes a rendering engine that maps JSON specifications from the agent to the project's own design system components for "generative UI."

Who it’s for

Angular developers building AI-powered applications who need a robust, signal-based chat surface that can integrate with LangGraph or other AG-UI compatible agent runtimes.

Highlights

  • Runtime Neutrality: Swap between different agent backends (e.g., LangGraph to AG-UI) without changing the chat components or templates.
  • Angular Signals Integration: Native use of Angular Signals for messages and status, removing the need for async pipes or zone.js.
  • Generative UI: Ability to render agent-generated JSON as custom design system components via a controlled registry.
  • Advanced Agent Support: Built-in handling for durable threads, human approvals (interrupts), subagents, and tool call progress.

Related

  • Project
    pax-beehive/paxm

    A memory adaptor for AI coding agents that provides a unified, cross-agent memory layer to store and recall project context and decisions.

  • Project
    pax-beehive/dsh-hub-cli

    A Node‑based CLI that lets you capture a DeepSeek Harness (runtime, ordered plugins, config patch, env‑var declarations) as an immutable, versioned *Preset Release* on the public DSH Plugin Hub. Others can fetch and apply the preset with a single command, getting exact plugin versions, load order, and runtime pinned. The tool supports planning, atomic staged installs, diff/upgrade/rollback, and integrates with AI agents via a safe plan‑apply workflow. All releases are content‑addressed (sha256), secrets stay local, and the CLI is MIT‑licensed.

  • Project
    oleksiijko/pmb

    A local-first memory system for AI coding agents that uses SQLite and MCP to store and recall project decisions, lessons, and constraints without cloud dependencies.

  • Project
    TetreesEX/TetreesAgent_EX

    Tetrees Agent EX is a Node.js SDK/CLI that lets developers search, quote, run, extend, and publish AI Packs on the Tetrees EX marketplace via the public MCP contract. It provides example scripts for buyer, builder, and seller flows, and keeps all credentials local.