Summon your army of AI agents
Switching between AI coding agents is chaos. ELVES gives them a shared workspace, shared memory, and a single place to ship from. Works with Claude Code & Codex!
ELVES creates isolated git worktrees for each task, allowing parallel AI agent work without interference. Users describe tasks in plain English, and the system spawns agents into dedicated workspaces with embedded terminals for real-time monitoring.
ELVES uses SQLite-backed persistent memory with full-text search to store decisions, context, and preferences. Memory relevance decays over time, keeping important information accessible while reducing noise.
ELVES provides a usage insights dashboard with real Claude Code telemetry including token usage by model, daily activity, project breakdowns, session quality metrics, and AI-generated narrative reports.
ELVES provides a WebGL-accelerated embedded terminal to watch agents work in real-time, then offers one-click merge with strategy selection (merge, rebase, squash) to review and ship completed work.
ELVES supports runtime-agnostic switching between Claude Code and Codex with the same interface, memory system, and worktree isolation, allowing users to test different engines without workflow disruption.