← back to ecosystem
app beta version 0.1.0 hex unreleased package jido_live_dashboard elixir ~> 1.17

Jido Live Dashboard

Real-time observability and debugging for Jido agents in Phoenix LiveDashboard

Beta support. Public iteration with faster feedback loops before final release shape settles.

Docs GitHub ★4 Issues
View package metadata source →

RELATED PACKAGES

AT A GLANCE

Zero-config integration — one function call adds all monitoring pages
Automatic telemetry capture for 13 Jido event patterns
ETS ring buffer for high-performance event storage
Trace correlation with trace_id/span_id linking

DEEP DIVE

Overview

Jido Live Dashboard is a real-time observability and debugging toolkit for the Jido agent ecosystem, built as a native extension to Phoenix LiveDashboard. A single call to JidoLiveDashboard.pages() adds four purpose-built dashboard pages covering system health, component discovery, runtime process inspection, and distributed trace analysis.

Purpose

Jido Live Dashboard serves as the observability layer for the Jido ecosystem. It answers: What components are registered? Which agents are running? What happened during that signal flow?

Major Components

Dashboard Pages

  • Home — System status, discovery catalog summary, running agents, trace buffer stats
  • Discovery — Tabbed browser for Actions, Agents, Skills, Sensors, and Demos
  • Runtime — Live AgentServer process table, WorkerPool status, InstanceManager stats
  • Traces — Telemetry event viewer with span hierarchy, measurements, and metadata

TraceBuffer

GenServer with ETS ring buffer capturing 13 Jido telemetry event patterns with trace_id/span_id correlation, configurable buffer sizes, and automatic pruning.

Runtime Introspection

Fault-tolerant functions for querying live Jido infrastructure: agent enumeration, state inspection, pool status, and discovery catalog queries.