$ show coder/xum
Xum (formerly Mux) is an open-source AI coding assistant multi-agent platform. Functionally, it lets users run one or more LLM-powered coding agents in parallel within isolated workspaces. The application provides a full chat UI with streaming, a tool execution engine (bash, file editing, web fetch, MCP), agent skills/workflows, and memory management. It supports multiple AI providers (Anthropic, OpenAI, Google, etc.) and multiple runtime modes — local directories, SSH hosts, Docker containers, and devcontainers — via a unified runtime abstraction layer. Architecture: The system is a hybrid monolith/dual-deployment. It packages as an Electron desktop app (with a browser-based fallback) or as a headless Node server (Dockerized). Typed IPC is handled by in-house oRPC over WebSocket/MessagePort. Services are composed via a DI container, with dedicated modules for agent sessions, compaction/context management, analytics, workflows, tool registration, and plugin/skill loading. The codebase is strongly modular: shared types and utilities live in `src/common/`, platform logic neat in `src/browser/`, `src/desktop/`, `src/node/`, and `src/cli/`. The VS Code extension in `vscode/` uses shared UI primitives. Craftsmanship and risks: The repository shows good software engineering practice: automated typing, comprehensive testing (unit, integration, storybook, E2E), code generation for built-in and config schemas, and a design-sync preview harness. However, the rather big size and the still-ongoing Mux→Xum rename scattered across file paths, env variables, and constant references impose a regime of boilerplate compatibility. Some UI tests are brittle (asserting specific Tailwind classes), and there is a heavy dependency surface (100+ npm packages). Overall the codebase is professional and well-structured.
The LLM identified 4 files worth highlighting for reuse:
Provides a createCoalescedReader function that wraps an async iterable and coalesces bursts of values into a single batch, waiting for a quiet period before yielding. Reusable in streaming, event batching, or consumption coalescing patterns.
A generic generator wrapper that pumps the underlying stream with incremental heartbeat values every, end with the final result. Keeps consumers alive when the producer becomes idle — useful for long‑running operations without producing final result for a while.
Provides family of stable‑reference hooks: useStableRecord, useStableMap, useStableArray call developer-side deep equality (Map/Record/Array) before returning the last unchanged value. Helps avoid unnecessary re‑renders in React when session data re‑packing is otherwise mismatched.
Implements groupDiffLines – a function that groups raw diff lines into logical chunks (headers, additions, deletions, context). Splits by change boundaries, preserves trailing context, and produces a structure view of inline diffs.