Scout
DocsBlogToolsContact
Browse Docs

Core Concepts

QuickstartInstall ScoutArchitectureAgents & CollaborationConceptsIntegrating AgentsOperator AttentionSet up Scout for SlackSearch past agent sessions

Walkthroughs

Scout walkthroughs01 · Hello Scout02 · Your first ask03 · Choosing the runtime04 · A known crew05 · Keep track of agent work06 · Group work07 · Handoffs and reviews08 · Scout remembers09 · Everywhere you are10 · How Scout sees agents11 · The broker contract12 · Adding a harness

Integrations

Scout for Grok BotHost integrationsMCP APIA2A and ACP readiness

Scout Chat

Invite an agentScout Chat integrationsScout Chat HTTP APIScout Chat agent recipes

Implementation

Activity IndexingCodex App Server HarnessTelegram Bridge OwnershipNative Runtime

Proposals

Proposal · Collaboration Agreements

Agent Paths

scout.jsonagent.mdagents.mdllms.txtllms-full.txtnav.jsoninstall.mdthis page.md
WalkthroughsIndex

Scout walkthroughs

Twelve walkthroughs in two tracks, from your own identity out to the broker contract.

View MD

Where you left off

0 of 12 read

Start at

01 · Hello Scout

You and the broker

Runnable start to finish: you type real scout commands against your own live broker and check the output as you go. The scope widens one ring at a time — from your identity and a single message, out to multi-agent handoffs, the mesh, and finally the broker contract itself.

If you have not installed Scout yet, do ../install.md↗ first. The first one assumes nothing else.

How each one reads

  • Goal — the capability you unlock.
  • You need — prerequisites. Most only need a healthy broker.
  • The idea — the mental model, in a few sentences.
  • Walk it — commands to run, with the output shape you should see. Output is abbreviated and representative; ids, times, and names will differ on your machine.
  • How to tell it worked — what a working setup looks like.
  • Push further — optional variations when you want more.
  • Next — where the ring widens next.

Commands are written for a POSIX shell. Anything in <angle brackets> is yours to substitute.

Track one: using Scout

For operators — you have coding agents and want Scout to coordinate them.

Hello ScoutYou and the brokerYour first askOne tracked unit of workChoosing the runtimeHarness, model, and session controlA known crewWorkers you can name and keepStaying in the loopMonitoring without pollingGroup workChannels, broadcast, rendezvousHandoffs and reviewsChained work across agentsScout remembersSearchable history and labelsEverywhere you areApp, phone, mesh, Slack

Track two: building on Scout

For developers — you want to integrate a harness, drive the broker from code, or understand what Scout owns.

How Scout sees agentsIdentity, sessions, observationThe broker contractRecords, routing, MCPAdding a harnessA worked end-to-end integration

The one rule

When a walkthrough disagrees with your machine, believe your machine and check scout doctor. These pages describe a healthy local suite; the fix for most surprises is in quickstart.md or scout doctor --fix.