| name | jido |
| description | Use when working with Jido 2.0 agent framework — defining agents, actions, plugins, signals, strategies, and the AgentServer runtime. Use when building multi-agent or AI-powered features in Elixir. |
This section is the authoritative reference for Jido's primitives. These pages aren't tutorials - they explain what each primitive is, why it exists, and how it fits with the others. Read them in order the first time through; each concept builds on the one before it.
The core model
Jido separates concerns that most agent frameworks collapse together. Actions are pure functions - validated, composable units of work that transform data. Signals are the universal message format, built on CloudEvents, that carry events and commands through the system. Agents are typed state structs with a behavior contract: pass in an action, get back updated state and directives. Directives are declarative descriptions of side effects - the agent never executes them directly. The runtime picks up those directives and executes them inside a supervised GenServer, keeping your domain logic deterministic and testable. Execution, the pipeline that actually runs actions, handles validation, chaining, retry, and compensation so callers don't have to.
Beyond the core, Jido has three cognitive pillars: Thread records what happened (the append-only interaction log), Memory stores what the agent currently believes and intends (the mutable cognitive substrate), and Strategy controls how actions execute. Sensors bridge external events into the signal layer, Plugins package reusable capabilities for composition across agents, and Persistence lets agents survive restarts through hibernate, thaw, and storage adapters.
Concepts (root files) — read in order
actions.md — Pure functions: validated, composable units of worksignals.md — CloudEvents-based message envelopes for commands & eventsagents.md — Typed state structs with a command interfacedirectives.md — Declarative side-effect descriptions returned by actionsagent-runtime.md — GenServer layer that executes directives & manages lifecyclesensors.md — Stateless modules bridging external events into signalsstrategy.md — Pluggable execution models for action processingplugins.md — Composable behavior bundles (actions, routes, state)execution.md — Action pipeline: validation, chaining, retry, compensationthreads.md — Append-only interaction logmemory.md — Mutable cognitive substrate (beliefs, goals)persistence.md — Hibernate/thaw and storage adapterstesting.md — Testing patterns for agents and actions
getting-started/
index.md — Getting started overviewinstallation.md — Installation & deps setupfirst-agent.md — Build your first agentfirst-llm-agent.md — Build your first LLM-powered agentelixir-developers.md — Orientation for Elixir developersnew-to-elixir.md — Orientation for developers new to Elixir
guides/
index.md — Guides overviewbuilding-a-weather-agent.md — End-to-end weather agent walkthroughdebugging-and-troubleshooting.md — Debugging patterns & diagnosticserror-handling-and-recovery.md — Error handling, retry, compensationpersistence-and-checkpoints.md — Persistence & checkpoint patternstesting-agents-and-actions.md — Comprehensive testing guidecookbook/chat-response.md — Chat response cookbook recipe
learn/ — step-by-step tutorials
index.md — Tutorials overviewfirst-workflow.md — Your first workflowai-agent-with-tools.md — AI agent with tool useai-chat-agent.md — Multi-turn chat agentmemory-and-retrieval-augmented-agents.md — Memory & RAG agentsmulti-agent-orchestration.md — Multi-agent orchestrationparent-child-agent-hierarchies.md — Parent-child agent hierarchiesplugins-and-composable-agents.md — Plugins & composable agentsreasoning-strategies-compared.md — Reasoning strategies comparedsensors-and-real-time-events.md — Sensors & real-time eventsstate-machines-with-fsm.md — State machines with FSMtask-planning-and-execution.md — Task planning & execution
reference/
index.md — Reference overviewbehavior-first-architecture.md — Behavior-first architecture designconfiguration.md — Configuration keys & optionsglossary.md — Term definitionsreq-llm-and-llmdb.md — ReqLLM & LLM DB integrationtelemetry-and-observability.md — Telemetry, metrics, observabilitywhy-not-just-a-genserver.md — Why Jido vs plain GenServer
ecosystem/ — package docs
jido.md — Core runtime packagejido_action.md — Action libraryjido_ai.md — LLM orchestration & reasoningjido_signal.md — Signal types & routingjido_memory.md / jido_memory_os.md — Memory subsystemsjido_chat.md / jido_chat_discord.md / jido_chat_telegram.md — Chat integrationsjido_mcp.md — Model Context Protocoljido_claude.md / jido_gemini.md / jido_bedrock.md — LLM provider adaptersjido_browser.md / jido_shell.md / jido_vfs.md — Tool packagesjido_workspace.md / jido_codex.md / jido_opencode.md — Dev toolsjido_cluster.md / jido_messaging.md — Distributed infrastructurejido_otel.md / jido_live_dashboard.md — Observabilityjido_character.md / jido_eval.md / jido_evolve.md — Specialized agentsjido_amp.md / jido_behaviortree.md / jido_runic.md — Execution strategiesjido_harness.md / jido_lib.md / jido_studio.md — Supporting librariesash_jido.md — Ash Framework integrationllm_db.md / req_llm.md — LLM client libraries
features/ — capability overviews
agents-that-self-heal.md — Self-healing agent patternsbeam-for-ai-builders.md — Why BEAM for AIbeam-native-agent-model.md — BEAM-native agent modelexecutive-brief.md — Executive summaryhow-agents-work.md — How agents work internallyjido-vs-framework-first-stacks.md — Jido vs other frameworksllm-support.md — LLM support & providersmulti-agent-coordination.md — Multi-agent coordinationobserve-everything.md — Observability featuresstart-small.md — Start small, scale uptools.md — Tool integration
build/ — implementation guides
index.md — Build overviewmixed-stack-integration.md — Mixed-stack integration patternsproduct-feature-blueprints.md — Product feature blueprintsquickstarts-by-persona.md — Quickstarts by role/personareference-architectures.md — Reference architectures
