Hermes series · Training 01

Get your first AI employee running on Hermes

Install an agent harness you own, plug in a brain you already pay for, and hire one agent that sorts your inbox before lunch. About forty five minutes, no coding.

🗂️ Email triage Read time ~14 min Build time ~45 min Last verified Sept 7, 2026
Level 1 · Use it

What a triage agent feels like on a normal day, and the one idea that makes every other AI decision easier. No setup. Start here.

Level 2 · Build it

Five steps on your own machine, with the real walls called out and a check after each one.

Level 3 · Under the hood

Which subscriptions actually work, what a fleet costs, and where to run it once your laptop is not enough. Optional.

Appendix

How many agents to build, what broke while we recorded this, a glossary, and the source docs.

One rule for this whole guide

Wherever we name a tool we use, you will see a 🔁 Swap box with alternatives. The entire reason this stack is worth learning is that it does not marry you to one company. Use whatever you already pay for.

Level 1 · Use it

What a triage agent feels like

One agent, one job. It reads your mail, sorts it, and interrupts you only when a human decision is genuinely required. The rest of the time you hear nothing, which is the whole point.

A normal Tuesday

  • 8:00am. Overnight mail is already sorted. Three things need you, eleven do not. You read the three.
  • Every fifty minutes. It wakes up, reads whatever arrived, labels it, and says nothing at all.
  • 11:30. Something needs a decision, so it messages you. One line: who, what they want, and by when.
  • All day. You never open your inbox to find out whether it needs you. That question stopped being your job.

The daily rhythm

The mechanism behind that day is called a heartbeat. You give an agent one standing instruction and an interval. Each time the interval passes and the agent is idle, that instruction fires as though you had typed it. Then the part that makes it livable: if nothing changed, it says so in one line and stops.

HEARTBEAT · EVERY 50 MINUTES 9:00quiet 9:50quiet 10:40quiet 12:20quiet 1:10quiet An email needs your attention 11:30 it speaks once · the other five passes cost almost nothing and say nothing
Silence is the feature. You spend attention only when there is something worth attending to.

Where the sorting ends up

Every message gets exactly one label, and you get one short report. Nothing is deleted and nothing is sent on your behalf, so the worst case on a bad day is a message filed in the wrong bucket.

inbox · 14 unread "Can you approve the invoice before Friday?" + 13 more you have not read 🗂️ Triage ONE REPORT · 1 NEEDS YOU to do 1 fyi 6 marketing 7 Approve invoice, due Friday everything else: filed, no action
Fourteen emails in, one decision out. The other thirteen are filed, never deleted.

Three ways it reaches you

An agent that only exists in a window on your desk is not an employee. Pick one channel and it can find you anywhere.

💬A private chat botEasiest to set up. Scan a code. 🧵Your work chatThreads, per topic. Tag it in. 📱A text messageFor "this needs you" taps. 🗂️ your triage agent
Same agent, whichever surface you already check. It only speaks when it has something.
🔁We use Telegram to start, because setup is a scanned code and nothing else. Alternatively: your work chat if that is where your day already happens, or a text message to your phone for the genuinely urgent taps. Setup for each is in Level 3.

The one idea: harness against model

This is the part that confuses people, and everything else follows from it.

The model is the brain. It reads, reasons and writes, and that is the whole job. It cannot remember yesterday, open your inbox, or wake up on its own. You rent it, and a better one ships every few months.

The harness is the body. It is the application that holds your memory, your connectors, your permissions and your schedule, and talks to the model on your behalf. Most chat apps you already use are a harness with one company's brain bolted in. An open harness lets you choose the brain, per agent, and change it later without rebuilding anything.

THE BRAIN · RENTED, SWAPPABLE any big model a cheap fast one an open model one on your Mac The harness yours to keep Memory Connectors Schedule Permissions Identity change the brain any time · keep everything below
You have been shopping for brains. The harness is the part you actually feel every day.

Why bother, if you already have a coding agent

Because they are different jobs, and you should keep both. A coding agent is your workshop: you open it, you drive, and it stops when you close the window. This is your staff: a roster of named agents with their own brains, tools and clocks, working while you are asleep and reaching you where you already are.

The three reasons

Use any model you want. Heartbeats, so it checks instead of you. And one agent per job, so each one does it properly.

Any model means the subscription you already pay for becomes the engine, and your cost per agent turns into a decision rather than a bill that arrives. Heartbeats mean "go check on it" stops being your job. Specialists mean shorter instructions, cheaper runs and noticeably better output than one assistant trying to do everything.

For a big picture person

You can stop reading here and still have the whole idea. Level 1 is the product. Everything below is for when you want it actually running on your own machine.

Level 2 · Build it

Build it

Five steps on your own machine. Nothing here needs the terminal except one optional prep command, and nothing here is code.

What you need before you start

  • A computer you can leave switched on. Mac or Windows both work. On a Mac you also get tools for your messages, notes and calendar.
  • An AI subscription the harness can sign into, or an API key. Which is which matters, and it is the single most expensive thing to get wrong. Read the brains table before you buy anything.
  • A package manager on your machine. If you already use a coding agent you almost certainly have this. If not, step 00 is thirty seconds.
  • An account with a connector service, so the agent can reach your mail. The free tier is genuinely enough to run this.
STEP 01 Install Download, double click. ≈ 5 min STEP 02 Give it a brain Sign in. No key needed. ≈ 2 min STEP 03 Give it hands One key, all the doors. ≈ 4 min STEP 04 Give it a job One prompt. ≈ 6 min install the body · rent a brain · open the doors · write the job description
The whole build. Everything after step 04 is refinement.
00

Prep, so the install does not stall

≈ 5 min · skip if you already use a coding agent

The harness pulls down a handful of open source tools it needs to operate your machine. If they are already there, install takes about five minutes. If they are not, it takes considerably longer, and it looks stuck while it works.

Terminal · macOS
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
Why this matters

A package manager is the hardware store your agent walks to. Mid job it realises it needs a tool it does not have, and with one installed it can go and fetch it rather than stopping and asking you. Version control is the other thing it reaches for constantly.

🔁We use Homebrew on macOS. Alternatively: winget or Chocolatey on Windows, or apt on Linux. Any package manager your agent can call from a terminal does the job.
✔ You'll know it worked when: asking your terminal for the package manager's version gets an answer instead of "command not found".
01

Install the harness

≈ 5 min, longer on a brand new machine
  1. Go to the project's own site and take the download for your operating system. It is the first link on the page.
  2. Run the installer. It walks you through permission levels and pulls down its starter tools.
  3. Launch it. The first open does a few more setup steps, then you land in the app.
The real wall

Two things, and both cost us time. First, check the address: searching for the project by name surfaces lookalike domains, so type the address from the official docs rather than trusting a search result. Second, the install looks frozen when it is not. On a fresh user account it sat for several minutes, reported failures on individual tools, then quietly installed them as it went. Let it finish before you decide it broke.

🔁We installed straight onto a laptop to keep the first run simple. Alternatively: install a container runtime first and run the harness inside a container, so the agent lives in a box with no reach into the rest of your files. Worth the extra step on a personal machine that has things on it you care about.
✔ You'll know it worked when: the app opens and asks you which model you want to use.
02

Give it a brain

≈ 2 min

The first question the app asks is which model to use. The cheapest good answer is a subscription you already pay for: some providers let a third party sign in on your behalf, which means no API key and no per token bill arriving later.

  1. Choose the sign in option for your provider rather than the API key option. It may sit under a "more providers" expander.
  2. It opens the provider's site. Log in and continue.
  3. It asks for a code, and the app shows you one. Paste it in, continue, then close the page.
  4. Back in the app it now says connected. That is your brain.
The real wall

Not every provider allows this, and the ones that do not get expensive fast, because the fallback is paying per token through an API key. This is the one decision on this page worth five minutes of reading first: see the brains table.

🔁We used a general purpose chat subscription for the walkthrough, because most members already have one. Alternatively: a different provider's subscription, a router subscription that gives you hundreds of models on one flat plan, a plain API key from any provider, or a model running locally on your own hardware for no ongoing cost at all.
✔ You'll know it worked when: the model picker in the chat window lists models you recognise.
03

Give it hands

≈ 4 min

Your agent now thinks but cannot touch anything. A connector service fixes that with one key that opens all the doors, rather than you wiring up each app by hand.

  1. Create an account with the connector service. Its free tier is metered by how often your agents call a tool, and it is generous enough that we have not run out.
  2. Connect your mail account, then sign in and approve the permissions.
  3. Label the connection while you are there. If you run several inboxes, add them all now: this is the moment multi account becomes easy instead of annoying.
  4. Ask it to connect your agent, and pick your harness from the list. It hands you a ready made prompt.
  5. Copy that prompt, paste it into your agent, and approve the commands as it runs. Under two minutes.
Why one key beats ten integrations

Every agent you create after this inherits the same tools, so you never do the wiring twice. And if you ever change harness or model provider, you move one key rather than reconnecting every app.

The real wall

Permissions. It is tempting to approve everything so the demo works, and for a read and sort job that is more access than you need. Most connector services have an advanced control that gives you read, write and delete separately per connection. Connect your mail read only for the first day, watch what it does, then widen it.

🔁We use a dedicated agent connector service because it is built for exactly this. Alternatively: the general purpose automation platform you may already pay for offers the same one key idea, or install the harness's own built in connectors directly if the app you need is on their list. All three give the agent the same reach.
✔ You'll know it worked when: you ask your agent what tools it has, and your mail is in the answer.
04

Give it a job

≈ 6 min

Start a fresh session first. New task, new session, always: a conversation carrying yesterday's context burns tokens for nothing.

Stay in your main profile. That agent acts as the recruiter here. It scouts, builds, trains and onboards the new hire, writing the job description, pinning the model and setting the tool access. Then paste the prompt from the Starter kit below and let it work.

The real wall

Letting a brand new agent loose on a live inbox. The prompt below ends with a dry run instruction on purpose: it shows you the labels it would apply to your last twenty messages and waits for your yes. Read that list properly. It is the cheapest possible moment to discover it thinks every newsletter needs your attention.

✔ You'll know it worked when: a second profile appears in your switcher with the name you gave it, and it reports a proposed label for each of your last twenty messages without having changed anything yet.
05

Verify, then put it on a clock

≈ 4 min
  • Read the dry run. Check the edge cases, not the obvious ones. A newsletter you actually chose to receive should be information, not marketing.
  • Inspect what it built. Open the new agent's profile and read its job description file. Your main agent wrote it. Tighten anything that reads vaguely, and add the rules it missed.
  • Check which model it picked. A sorting job should be on something cheap and fast. If it pinned an expensive model, change it now.
  • Turn it loose for real. Switch into the new profile and tell it to run the pass properly. The labels appear in your mail.
  • Then give it a heartbeat, in plain words, for example: run this every fifty minutes and stay quiet unless something new needs me.
✔ You'll know it worked when: your labels exist in your mail client, and an hour later a second pass has happened without you asking.
Starter kit

Starter kit

Three copy blocks. The first hands the whole build to your own coding agent, the second hires the triage agent, and the third is the blank job description to reuse for every hire after this one.

1. Hand the whole build to your agent

If you would rather not click through the steps yourself, fill in the four blanks at the top and paste this into the coding agent you already use. It reports after each step instead of working silently.

Install prompt · fill in the blanks
# FILL THESE IN FIRST
#   [OS]        = macOS or Windows
#   [PROVIDER]  = the AI provider whose subscription you already pay for
#   [MAIL]      = the mail account you want triaged
#   [CHANNEL]   = where you want to be pinged (chat bot, work chat, text)

You are helping me install an open source agent harness on my [OS] machine and
get one working agent running. I am not a developer. Explain each step in plain
words, and stop and report after each numbered step rather than doing all of it
silently. If something fails, tell me what you saw before you retry.

1. Check whether I already have a package manager and version control installed.
   If not, install them and confirm.

2. Walk me through downloading and installing the harness's desktop app for [OS].
   Tell me where to get it from the official documentation, not a search result.
   Warn me before the step that usually looks frozen.

3. Help me connect [PROVIDER] as the model, using subscription sign in rather
   than an API key if that provider supports it. If it does not support sign in,
   stop and tell me before I spend anything, and show me the options.

4. Help me connect a tool connector service so the agent can reach [MAIL].
   Set the mail permission to read only for now. Confirm what access it has.

5. Confirm the agent can list its own tools and see [MAIL].

6. Set up [CHANNEL] so the agent can reach me when I am away from my desk.

Do not create any agents yet. Stop when step 6 is confirmed and tell me what
you set up, then I will give you the job description separately.
✔ You'll know it worked when: your agent reports six confirmed steps, and asks you for the job description.

2. Hire the triage agent

This is the prompt from the walkthrough. Change the two bracketed values and paste it into your main profile.

Agent prompt · paste into your main profile
Set up a new specialized agent for me, then give it its job.

Create a new profile called "triage" with this identity:
  Name: Triage
  Role: Email chief of staff for [YOUR NAME] at [YOUR COMPANY].
  Job:  Keep my inbox free of decisions I haven't made yet.
        Read new mail, label it, surface only what needs me.

Configure it:
  - Model: pin a fast, cheap model. This job is sorting, not reasoning.
  - Access: use my existing mail connector.
  - Skills: only what it needs for email. Turn the rest off.
  - Create the labels TO DO, FYI and MARKETING if they don't exist.

Write its job description file so it works exactly like this:

1. Each run, read unread mail from the last 24 hours.
   Skip anything already labeled.

2. Apply exactly one label per message:
     TO DO      someone is waiting on a reply, a decision, or money from me
     FYI        I should know it, no action: receipts, confirmations, CCs,
                reports, newsletters I chose to get
     MARKETING  cold outreach, promos, blasts, anything selling to me
   When it is genuinely a coin flip, label it TO DO and say so.
   A missed reply costs me more than a wrong FYI.

3. Never archive, delete, reply, forward or send. Labeling only.

4. Then write me one report, in this shape:
     TO DO (n)      one line each: who, what they want, deadline if stated
     FYI (n)        one line total
     MARKETING (n)  count only
   If TO DO is empty, say "Nothing needs you." and stop.
   Do not narrate an empty inbox.

Then set its heartbeat:
  every 50 minutes, run the pass above and stay silent unless
  something new landed in TO DO.

Before you touch my inbox: do one dry run. Show me the labels you
would apply to the last 20 messages and wait for my yes.

Why each line is in there

Read this once and you can write the next agent yourself. These seven choices are the transferable part, not the email specifics.

The lineWhat it buys you
Give it a nameThe agent gets its own folder, its own memory and its own command. Names matter for you too: you need something to call when you tell one agent to hand work to another.
Pin a cheap modelSorting subject lines is not reasoning. This is the single line that makes it affordable to leave running all day.
Exactly one labelForces a decision. Multi label triage quietly recreates the pile you were trying to clear.
Name the expensive mistakeTelling it that a missed reply costs more than a wrong filing means it knows which way to err when it is unsure.
Never send or deleteKeeps the blast radius at zero while you are still learning to trust it. Grant more powers once it has earned them.
Silent by defaultWithout this line your agent becomes a second inbox, which is worse than the first one.
Dry run firstYou approve its judgement on twenty real messages before it is allowed to write anything.

3. The blank job description

Every agent after this one starts here. Fill in the six sections and you have a new hire; the shape is what keeps them predictable.

Job description template · reuse for every agent
### Identity
Name: [AGENT NAME]
Role: [ONE LINE. Who they work for and what they own.]
Job:  [ONE SENTENCE. The outcome, not the activity.]

### On every run
1. [What to read, and how far back.]
2. [What to skip, so it does not redo finished work.]
3. [The decision it makes, with the exact options allowed.]
4. [Which way to err when it is genuinely unsure, and why.]

### Never
- [The irreversible actions it must not take.]
- [Anything that speaks to another human as me.]

### Report
Write one report in this shape:
  [SECTION] (n)   [one line each: who, what, when]
  [SECTION] (n)   [one line total]
If there is nothing that needs me, say "[YOUR QUIET PHRASE]" and stop.
Do not narrate an empty result.

### Schedule
Run [HOW OFTEN] and stay silent unless [WHAT MAKES IT WORTH INTERRUPTING ME].

### Before you touch anything live
Do one dry run on the last [N] items. Show me what you would do and wait
for my yes.

Setup checklist

  • Package manager and version control installed.
  • Harness installed and opening without errors.
  • A brain connected, and you know whether you are on a subscription or paying per token.
  • Mail connected, read only to start, and labelled in your connector.
  • A fallback model set, so one provider having a bad day does not stop your work.
  • Dry run read and approved by you, not skipped.
  • A heartbeat set, and one channel where it can reach you.
Level 3 · Under the hood

Brains, cost, and where it runs

Optional. This is the section that saves you money, and the first table is the one thing we would ask you to read before you buy anything.

Which subscriptions a harness can sign into

Providers split into two camps, and the difference is the whole cost story. Sign in means the harness logs into a plan you already pay for, so there is no separate bill. API key means you pay per token on top of whatever else you pay, and an agent that runs all day is a meter that never stops.

ProviderHow it authenticatesWhat you need to know
OpenAI
via the Codex provider
Sign inDevice code login with your existing chat subscription, and it runs the Codex models. This is what we used in the walkthrough. How much your plan tier allows is not documented, so expect to discover your own ceiling.
xAI
Grok
Sign inBrowser sign in, no key needed, on the premium consumer tiers. Strong at using tools, which matters more for agent work than raw benchmark scores.
GitHub CopilotSign inThe same device code flow as the Copilot command line tool. Easy to overlook if you already pay for it for coding.
Anthropic
Claude
Top tier onlySign in works on the Max plan with purchased extra usage credits. It is not available on the mid tier plan, which has to use an API key instead. Worth checking your own plan before assuming either way.
Everything else
routers, Gemini, DeepSeek, Qwen, NVIDIA and many more
API keyDozens of providers are supported, all with a key in a local config file. Fine for occasional heavy thinking, expensive for anything on a schedule.
A model on your own hardwareFreeNeeds a capable machine. Costs nothing per run, your data never leaves the building, and that unlocks work which cannot legally go to a third party.
Check this yourself before you buy

Which plans allow third party sign in is a commercial policy, and it moves. Two of the rows above changed shape in the months before this page was written. Read the harness's own provider documentation on the day you set this up rather than trusting any guide, including this one.

🔁We run one provider as the default and a second as backup. Alternatively: a single router subscription covers hundreds of models on one flat plan, which is simpler to reason about but bills as API usage, so a busy fleet burns through it faster than you expect. Or go the other way and run a local model as the default, with a paid provider reserved for the hard jobs.

What a small fleet actually costs

  • The entry level plan will frustrate you. Consumer tiers have rolling usage windows, and one agent on a schedule can exhaust one in an afternoon. You will spend more time waiting for a window to reset than you save.
  • A mid or top tier plan is the honest starting point. One of those carried five or six agents doing different jobs for us without running dry.
  • Budget more for your first week than your tenth. Setting an agent up costs far more than running it. You are correcting it, and a good harness rewrites its own job description each time you do, so the same mistake costs less next time.
  • Match the model to the job, not to your ambition. The single biggest lever on this page is putting sorting work on a cheap model and keeping the expensive one for work that genuinely reasons.
The pattern worth stealing

Build the agent on the smartest model you have. Once it knows its job, drop it to the cheapest one that can still do it.

Setup is where reasoning matters, and execution usually is not. Our triage agent was configured with a top model and left running on a small one. The counterweight is worth stating too: during setup, a cheap model on a hard job costs more in your wasted time than it saves in tokens.

Settings worth changing on day one

🛟

Set a fallback model

If you hit a usage limit or your provider has an outage, the harness switches to a backup instead of stopping. Your work is about to depend on these agents, so do not let one provider's bad day become yours.

🎛️

Per job model overrides

You can point one kind of work at one model and another elsewhere, and some harnesses will bounce a hard task between two. Ignore all of it on day one. One default is fine.

🧠

Memory budget

Raise it and the agent remembers more, at some cost to your machine. Third party memory providers plug in here too and extend it much further, which is worth a look once you are settled.

🔌

Where it runs

Local, hosted, or your own remote machine. You can change this later and keep the same desktop app in front of you, which is why starting local costs you nothing.

🧰

Tools, per agent

Leave everything off that this agent does not need. Fewer tools is not a limitation, it is how you stop an agent reaching for the wrong one.

🗂️

Sessions or a named roster

Some harnesses let profiles become named bots with group chats and handoff between them. We prefer plain sessions. This is preference, not correctness.

Where to run it, once your laptop is not enough

Start local. You will know the moment you have outgrown it: you are away from your desk, you want to ask your agent something, and the machine it lives on is shut.

Option 01 · start here

Your own machine

Download and go. Runs only while the machine is awake. On a Mac you also get your messages, notes and calendar as tools, and a small always on desktop left plugged in at home makes a capable little server.

Option 02

Hosted for you

The people who make the harness will run it, so there is no server to maintain and it is reachable while you are out. Check whether the hosted product is still in preview before you put client work on it.

Option 03

A rented remote machine

A real computer, which is the whole point: you can install actual applications for your agents to use, which a hosted product will not let you do. Providers keep uptime high, so it beats a machine at home that shares your power supply.

🔁We started on a laptop and later moved to a rented remote machine. Alternatively: a small always on desktop at home if you want the phone and calendar tools, the harness maker's own hosted option if you never want to touch a server, or any mainstream virtual machine host from about two gigabytes of memory upward. The harness is the same in all three; only the address changes.

Giving an agent its own identity

If an agent is reading your inbox, it needs access to yours and that is that. But the moment an agent starts communicating with other people, give it its own identity: its own mail address, and its own number if it texts or calls.

Two reasons, and the second one surprised us. It stops the agent speaking as you, which is a real liability question. And it puts you in the right frame of mind: you send an assistant a request, you do not think their thoughts for them.

🔁We give outbound agents their own mailbox on our own domain. Alternatively: services now exist that create mailboxes designed for agents, which notify the agent the moment mail arrives instead of making it check on a timer, and there are equivalents for giving an agent a spending method. Or simply keep every agent read only and inbound, which is the safest starting point and the one we recommend for your first month.
Appendix

Appendix

Not just one agent

Triage is the easiest first hire because everyone has the problem. The same four part shape builds any of these, and the job description template in the Starter kit is the only file that changes.

🗂️

Triage

Labels the inbox, escalates only decisions. What this guide builds.

📇

Setter

Watches for new enquiries and drafts the first reply. Never sends without you.

📰

Correspondent

Reads your industry overnight and files one brief you can skim over coffee.

🧾

Bookkeeper

Flags uncategorised spending and anything that looks like a duplicate charge.

How many agents, and how specialised?

There is no single right answer, but there is a trap. You will see people online with fleets of a hundred agents. That is payroll bloat, and it creates exactly the problem a large company has, where nobody can find out who owns what.

Our view

Build a flat team of broadly capable agents. A small team with big output.

Think of a T shaped hire: deep in one thing, capable across the neighbouring ones. An agent that handles search visibility and can also write copy and touch a website is really a content manager, and that is the right size of job. We recently collapsed our own fleet down to role level agents and the output got better, not worse.

You also do not need to talk to all of them. Run several, but speak to one: a chief of staff that assigns work to the others and reports back when it is done. Most harnesses have the plumbing for this already, including a board view where you can watch the handoffs happen if you want to.

What broke while we recorded this

We left these in on purpose. All four are normal, and knowing them in advance is the difference between a slow afternoon and a lost one.

  • The install sat still for several minutes. A brand new user account meant every supporting tool was fetched from scratch. It even reported failures on individual tools and then installed them anyway as it went. Doing step 00 first avoids most of this.
  • The first dry run changed nothing, which looked like failure. It is meant to. It proposes labels and waits. Do not go looking for a bug in your mail client.
  • The cheap model was visibly slower on work outside its job. Once triage was pinned to a small model, asking it to draft a message took noticeably longer. That is the trade you chose, and it is the right trade for sorting.
  • You approve a lot of commands in the first hour. Familiar if you already use a coding agent, and it settles down quickly as the agent learns what it is allowed to do.

Plain English glossary

TermWhat it actually means
ModelThe brain you rent. It reads, reasons and writes, and it forgets everything the moment the conversation ends.
HarnessThe application around the brain. It holds the memory, the tool access, the permissions and the schedule. This is the part you own.
AgentOne harness profile with a job description, a pinned model and its own memory. An employee, not a prompt.
HeartbeatA standing instruction plus an interval. The agent wakes, does the job, and stays quiet unless something needs you.
Job description fileA plain text file holding an agent's identity and rules. Editing this file is how you manage the agent.
ConnectorThe bridge between an agent and a real application, so it can read your mail rather than talk about mail.
Dry runThe agent shows what it would do and waits for your yes. The cheapest safety feature there is.
TokenThe unit AI usage is measured in. Roughly a word fragment. Relevant because an agent on a schedule spends them while you sleep.

Your homework before the next one

  • Install it and connect one brain. That is the whole first win, and nothing else counts until it is done.
  • Pick the ten minute task you resent. Not the impressive one. The annoying one. That is your first agent.
  • Give it a heartbeat and let it run for two days. Then bring what it got wrong to the community, because the wrong answers are how a job description gets sharp.
  • Post your roster. Two agents is a team. Seeing somebody else's is how you find your third.

Source documentation

Stuck?

Ask in the community. This is the first guide in a series, and the next ones get built from the questions you bring, so tell us where you got blocked and what you want an agent to do next.

Agentic Society · Hermes series · Training 01

Last verified Sept 7, 2026. These systems improve themselves, and the products move underneath them, so treat screen names and plan rules as current as of that date rather than permanent. Where this page and a vendor's own documentation disagree, believe the vendor.