Examples Catalog

The repo includes a complete, tested example suite at apps/examples/. Every example exports a run() function, can be executed standalone with bun run, and runs in CI. The fastest way to learn this framework is to copy one of these and tweak it.

Directory at a glance

• Directoryapps/examples/

  • Directoryfoundations/ the minimum builder chain → memory → composition

    • …
  • Directorytools/ built-in tools, MCP servers, dynamic registration

    • …
  • Directoryreasoning/ strategies + model-adaptive context profiles

    • …
  • Directorytrust/ identity, guardrails, verification

    • …
  • Directorymulti-agent/ A2A protocol, orchestration, dynamic spawning

    • …
  • Directorystreaming/ token streaming + SSE endpoints

    • …
  • Directoryintegrations/ Next.js · Hono · Express adapters

    • …
  • Directorygateway/ persistent autonomous agents

    • …
  • Directorymessaging/ Signal + Telegram via MCP

    • …
  • Directoryinteraction/ 5 autonomy modes

    • …
  • Directoryadvanced/ cost · observability · self-improvement · eval

    • …
  • Directorydemos/ standalone showcase apps (not part of the numbered/CI catalog)

    • …
  • index.ts CLI runner — bun run index.ts --filter <category>
  • README.md

Quick Start

git clone https://github.com/tylerjrbuell/reactive-agents-ts
cd reactive-agents-ts/apps/examples

# Run all offline examples (no API key needed)
bun run index.ts --offline

# Run a specific example
bun run src/foundations/01-simple-agent.ts

# Run or list a standalone demo through Rax
rax examples list
rax examples demo island-sim --provider ollama --model cogito:14b --port 3007

# Run all examples that match a category
bun run index.ts --filter foundations

---

Foundations

The shortest path from bun add reactive-agents to a working agent.

#	Example	What it shows
01	simple-agent	The minimum builder chain — provider, build, run
02	lifecycle-hooks	Intercept any of the 12 phases with before / after / on-error
03	multi-turn-memory	agent.session() for conversational memory
04	agent-composition	Compose specialized agents into pipelines
05	agent-config	Agent-as-data: serialize → JSON → reconstruct
06	composition	pipe(), parallel(), race() functional combinators

Tools

Built-in tools, MCP servers, and runtime tool registration.

#	Example	What it shows
05	builtin-tools	web-search, file-read, code-execute, etc.
06	mcp-filesystem	MCP filesystem server via .withMCP()
07	mcp-github	MCP GitHub server
—	mcp-oauth	OAuth 2.1 client_credentials against a remote MCP server via .withMCP({auth})
—	dynamic-registration	agent.registerTool() / unregisterTool() at runtime

Reasoning

The reasoning strategies and model-adaptive context profiles.

#	Example	What it shows
19	reasoning-strategies	ReAct, Reflexion, Plan-Execute, Tree-of-Thought, Adaptive
20	context-profiles	Local / mid / large / frontier tier tuning

Trust & Safety

Identity, guardrails, and verification.

#	Example	What it shows
11	identity	Ed25519 certificates, RBAC, delegation
12	guardrails	Pre-LLM injection, PII, toxicity blocking
13	verification	Semantic entropy, fact decomposition, NLI

Multi-Agent

A2A protocol, orchestration, and dynamic sub-agent spawning.

#	Example	What it shows
08	a2a-protocol	Agent Cards, JSON-RPC server/client, SSE
10	dynamic-spawning	.withDynamicSubAgents() — model-driven delegation

Streaming

Token streaming, SSE endpoints, and abort signals.

#	Example	What it shows
23	token-streaming	agent.runStream() AsyncGenerator with IterationProgress events
24	streaming-sse-server	One-line AgentStream.toSSE() HTTP endpoint

Web Framework Integrations

Drop-in adapters for popular Node/Bun servers.

#	Example	What it shows
25	nextjs-streaming	Next.js Route Handler with useAgentStream
26	hono-agent-api	Hono on Bun with SSE streaming
27	express-middleware	Express middleware pattern

Gateway

Persistent autonomous agents with heartbeats, crons, and webhooks.

#	Example	What it shows
22	persistent-gateway	Adaptive heartbeat + cron scheduling + webhook ingestion
25	hn-gateway-monitor	Real-world monitor: 24/7 Hacker News watcher with policy budgets
26	gateway-chat-mode	Per-sender SQLite session history + episodic context injection

Messaging

Connect agents to Signal and Telegram via MCP.

Example	What it shows
signal-telegram-hub	Multi-channel message hub bridging Signal + Telegram + the agent

Interaction

Autonomy modes and approval gates.

#	Example	What it shows
21	interaction-modes	Autonomous · Supervised · Collaborative · Consultative · Interrogative

Advanced

Cost tracking, observability, self-improvement, evaluation.

#	Example	What it shows
14	cost-tracking	Complexity routing + budget enforcement + dynamic pricing
15	prompt-experiments	A/B testing prompts via the prompt library
16	eval-framework	LLM-as-judge scoring with frozen-judge isolation
17	observability	Distributed tracing + metrics dashboard + structured logging
18	self-improvement	Cross-task strategy outcome learning
20	compose-harness	Composable harness pipeline
22	durable-hitl	Durable approval gates — run() pause → approveRun/denyRun + onApproval

Demos

Standalone showcase apps under apps/examples/src/demos/ are not part of the numbered index.ts catalog above (they have no run() export). Each has its own usage notes and focused tests where provided.

Example	What it shows
halopedia-agent	Full-featured interactive research agent: 7 custom domain tools (search, page/section/table fetch, cross-entity compare, link-graph explore), persistent cross-session memory (.withMemory() + find/recall/relate meta-tools), .withToolIntent() for domain-tuned chat routing, agent.session({ onOverflow }) for summarize-on-overflow history compaction, and verifyCitations: true to catch uncited/fabricated sources — run bun run apps/examples/src/demos/halopedia-agent.ts for a live Halo-lore REPL (requires Ollama)
local-vs-frontier	Same builder chain run twice — a local 4B model and a frontier model complete the same tool-using task, only the provider/model line changes (see demos/README.md for GIF recording)
canonical-chat-session	Reference agent.session() conversational loop
durable-resume	Crash-resume via .withDurableRuns() outside the eval-scoped 22-durable-hitl example
island-survival	Deterministic survival simulation with optional local world generation and camp-batched System One decisions; provider failures fall back to scripted moves

---

Running Patterns

# Offline only (test provider — no API keys needed)
bun run index.ts --offline

# Filter by category
bun run index.ts --filter reasoning

# Single file, standalone
bun run src/foundations/01-simple-agent.ts

# List or launch a standalone demo
rax examples list
rax examples demo island-sim --provider ollama --model cogito:14b --port 3007

# All examples (requires ANTHROPIC_API_KEY etc. in .env)
bun run index.ts

Every example is exercised by the test runner on every PR — if it’s listed here, it works.

Next steps

• Quickstart — set up your own project from scratch
• Common Builder Stacks — copy-paste chains organized by use case
• Choosing a Stack — match provider · model · memory to your workload