Preparatory homework: get ready to build your AI mail assistant (3–8 November)

On this page

The same page online, in parts: https://huettner.io/abp/ws2/hw/

What you will build in Workshop II

In Workshop II (9, 16 and 23 November), you will build an AI assistant for a real mailbox.

Your goal for 9 November

At the end of the first morning, this works on your laptop:

  1. You will write a mail from your SKKU mail account to your test Gmail account.
  2. Your AI assistant will sort the mail and draft Kaminski's answer.
  3. You will read the draft and say OK.
  4. The answer will arrive in your SKKU inbox.
  1. Your SKKU accountyou write to your test Gmail account
  2. Your test Gmail accountthe mail arrives
  3. n8npicks it up, checks it
  4. Langflow: Triagedecides: reply
  5. Langflow: Draft Replywrites in Kaminski's words, checks it
  6. n8nchecks the rules, asks you
  7. Yousay OK
  8. Your SKKU inboxthe answer arrives, from your test Gmail account
Langflow thinks. n8n does the work. You say OK.

So you work with two mail accounts: your SKKU mail account, which you already have, and a new test Gmail account, which you make in step 2.

Two ideas you know from Yanbo's workshop
  • Decomposition. Your AI assistant is not one big AI that does everything. The job is cut into small steps, and each step has one task: n8n picks up the mail, Triage only sorts it, Draft Reply only writes the answer, n8n checks the rules, and you decide. Small steps are easy to test and to fix. On 9 November you will test Triage on its own first.
  • Guardrails. Every step is checked before the next one acts. n8n looks at each new mail first: it hides personal numbers, and a mail that tries to trick your AI assistant goes to you instead. Langflow checks its own answer: in step 8 you will see "checks_ok": true. And nothing leaves without your OK.

This homework: three parts, and one optional

The homework gets your laptop ready for 9 November. It has three parts, and one optional:

Part Steps What you do
A 1–7 Set up your test mailbox: update kingo-pod, make the account, copy Kaminski's mails into it
B 8 Test Langflow: load the course flows, let Triage sort a mail
C 9–13 Set up n8n and test it: logins, mail search, a test mail from your SKKU account
Optional 14 Look into the database: see Kaminski's mails and your test mail as rows, in CloudBeaver
Gmailyour test Gmail accountPart A · steps 1–7
Mailkingo-pod's service: copies the mails, masks addressesPart A · steps 1–7
PostgreSQLevery mail as a rowPart A · steps 1–7
n8ndoes the workPart C · steps 9–13
Langflowthinks: sorts and draftsPart B · step 8
Part A · steps 1–7
Part C · steps 9–13
Part B · step 8
What this homework sets up on your laptop: the mails go from Gmail into your database, and n8n and Langflow work with them there.

Before you start

Part A · Set up your test mailbox

Gmailyour test Gmail accountPart A · steps 1–7 · you are here
Mailkingo-pod's service: copies the mails, masks addressesPart A · steps 1–7 · you are here
PostgreSQLevery mail as a rowPart A · steps 1–7 · you are here
n8ndoes the workPart C · steps 9–13
Langflowthinks: sorts and draftsPart B · step 8
Part A · steps 1–7 · you are here
Part C · steps 9–13
Part B · step 8

1. Update kingo-pod: get the new part called 'Mail'

kingo-pod is the course software on your laptop. Langflow and n8n run in it. This update adds a new part called Mail. It connects your test Gmail account to kingo-pod. Later, it also sends the answers you approve.

Open a terminal (on a Mac: the Terminal app; on Windows: the Ubuntu app) and run this one line:

cd ~/kingo-pod && git checkout -- . && git pull --ff-only && ./kingo backup && ./kingo update && ./kingo mode abp

Near the end, you should see a line that starts with Stack updated — kingo-pod 1.4.3 (or a higher number).

This can take a few minutes. You don't have to wait: go on with steps 2 to 5 in your browser. The update keeps running in the terminal.

In case you care about what this does
  • && just connects the commands: each one starts when the one before it worked. If one fails, the line stops there.
  • cd ~/kingo-pod goes to your kingo-pod folder.
  • git checkout -- . undoes changes in your kingo-pod folder. For Workshop I, many of you added nemoguardrails to langflow/requirements.txt. That change stopped every update since. kingo-pod now has nemoguardrails built in.
  • git pull --ff-only gets the newest version.
  • ./kingo backup saves your Langflow flows and n8n workflows first.
  • ./kingo update builds the new version.
  • ./kingo mode abp starts the parts we need: Langflow, n8n, the PostgreSQL database, CloudBeaver and a service called Mail.

Stuck? Help for step 1

2. Make your test Gmail account

  1. Go to https://accounts.google.com/signup. Google will ask for your phone number.
  2. Use this pattern for the address: n8n.<givenname>.<familyname>@gmail.com
    • For example, for Min-ji Kim: n8n.minji.kim@gmail.com (small letters, no spaces, no hyphens)
    • If the address is taken, add a 2: n8n.minji.kim2@gmail.com
  3. When Google suggests addresses, click Create your own Gmail address. Type the part before @gmail.com.

This account is only for the course. Do not use your private account or your SKKU account: the AI assistant you build will read and send mail in this account.

3. Set your test Gmail account to English

In Gmail, signed in with your test Gmail account: gear icon → See all settings → General → Language: English (US) → Save Changes (at the bottom of the page). Do this before step 6: then the folders in your test Gmail account have English names, as in this guide.

4. Turn on 2-Step Verification (needed for step 5)

Go to https://myaccount.google.com/security → 2-Step Verification → turn it on. Choose text messages, the Google prompt, or an authenticator app. A security key alone does not work for step 5.

5. Make an app password for kingo-pod

  1. Go to https://myaccount.google.com/apppasswords.
  2. Type a name for it, for example kingo mail, and click Create. The name does not matter; it only helps you find the password on this page later.
  3. Google shows 16 letters, only once. Copy them.

This is not your Google password. It is a second password, only for kingo-pod: with it, kingo-pod can read your test Gmail account and send the answers you approve. You can delete it any time on the same page.

Stuck? Help for step 5

6. Connect your test Gmail account to kingo-pod

First check that step 1 is done. In the terminal, in your kingo-pod folder, run:

./kingo status

You should see Mail in the list. Then run:

./kingo mail

It asks four questions:

  1. The address of your test Gmail account. Type the full address, with @gmail.com, for example n8n.minji.kim@gmail.com.

  2. The app password from step 5. You will not see it while you type.

  3. Other addresses kingo-pod may send mail to. Type your SKKU address. On 9 November you will write from your SKKU address, and your AI assistant's answers go back to it.

    Why kingo-pod asks for these addresses

    kingo-pod sends mail only to your test Gmail account and the addresses you type here. A mail to someone in Kaminski's mails lands in your own test Gmail account instead. A mail to any other address is not sent at all. So neither your AI assistant nor a mistake can send mail to anyone else.

    Forgot your SKKU address, or want to add another one later? Run ./kingo mail again. Every question now shows your last answer in brackets. Press Enter to keep it, also at the app password, and type all addresses at question 3.

  4. The sender name. Press Enter. That takes the name in the brackets, Vince J Kaminski: your AI assistant answers in his name.

kingo-pod then checks the login. It sends nothing. You should see a line that starts with OK. Folders:.

Below it, kingo-pod says how many of Kaminski's mails your test Gmail account already holds, and asks a question:

Your test Gmail account holds 0 of 3,244 of Kaminski's mails.
Upload them now? [Y/n]:

Your account is new, so it holds none of them. The question is the start of step 7.

Stuck? Help for step 6

7. Copy Kaminski's mails into your test Gmail account

Your AI assistant will search Kaminski's old mails for facts when it drafts an answer. The update in step 1 brought all 3,244 mails onto your laptop, inside kingo-pod. Now you copy them from kingo-pod into your test Gmail account.

  1. The terminal still shows the question from the end of step 6: Upload them now? Press Enter. That means yes.

    After a short while, you see a line that starts with Mailbox: and ends with Uploading starts now.

    Did the question not come, or did you answer no? Run this in the terminal, in your kingo-pod folder. It uploads the mails too:

    ./kingo mail seed
    
    What happens

    In a few seconds, kingo-pod puts all mails into its service called Mail, on your laptop. Mail keeps this copy: with it, Mail knows what is new on each side. From there, kingo-pod copies the mails into its PostgreSQL database and uploads them to your test Gmail account, round by round. In the end, every mail is in three places: in Gmail, in Mail, and in PostgreSQL, where n8n and your AI assistant read it.

  2. The command is done, but the upload goes on inside kingo-pod for about 30 minutes, sometimes longer. Be patient. Keep kingo-pod running and your laptop awake.

  3. Watch the mails arrive. They come step by step over the 30 minutes, so the numbers in Gmail grow while you watch. Open Gmail in your browser, signed in with your test Gmail account. When the upload is done, you see:

    • Sent (in the menu on the left): the mails Kaminski wrote, about 870 conversations.
    • All Mail (in the menu on the left, under More): about 1,600 conversations.
    • Inbox: stays empty for now. That is right: new mail comes in step 12 and in the workshop.
    Why not 3,244?

    Gmail counts conversations, not mails: a mail and all its answers are one conversation. So Gmail shows smaller numbers, but all 3,244 mails are there.

    To count the mails, ask kingo-pod: run ./kingo mail status. It shows each folder twice, here and at Gmail. When the numbers are the same in every row, the upload is done. (No such rows? Run ./kingo update first.)

  4. You don't have to wait for the upload: go on with Part B now. The upload keeps running in the background.

Good to know:

Stuck? Help for step 7

Part B · Test Langflow

Gmailyour test Gmail accountPart A · steps 1–7 ✓
Mailkingo-pod's service: copies the mails, masks addressesPart A · steps 1–7 ✓
PostgreSQLevery mail as a rowPart A · steps 1–7 ✓
n8ndoes the workPart C · steps 9–13
Langflowthinks: sorts and draftsPart B · step 8 · you are here
Part A · steps 1–7 ✓
Part C · steps 9–13
Part B · step 8 · you are here

8. Test Langflow: load the course flows and let Triage sort a mail

This part checks that Langflow works with the OpenRouter key Yanbo gave you, and with the flows of the course.

We prepared a Langflow flow for you: Triage (Enron). It reads one mail and decides what to do with it: reply, calendar, forward or archive. On 9 November, your AI assistant will use this flow for every new mail. Now you test it by hand.

Open Langflow: http://localhost:7860.

  1. Check your key. Langflow needs the OpenRouter key Yanbo gave you. Langflow keeps keys as global variables: you enter a key once, and every flow can use it.

    • Click your profile icon (top right) → Settings → Global Variables.
    • Is there a variable called OPEN_ROUTER_API_KEY? Then go on with step 8.2.
    • If not, click Add New. Name: OPEN_ROUTER_API_KEY. Type: Credential (then Langflow hides the key on screen). Value: your key. Save.
    • More help: Langflow: Global variables
  2. Get the course files. Download ws2_hw.zip and unzip it. You get a folder ws2_hw with two folders inside (langflow and n8n) and ws2_hw_homework.html: this homework as one page, to read without the internet. Every name starts with ws2_, for Workshop II. You need them now and in Part C.

  3. Load the Langflow flows. Make a new project called WS2 - Mail Assistant (Enron). Open the folder ws2_hw/langflow (from step 8.2) next to your browser. Select all four files in it and drag them onto the page of your new project, where its flows are listed (not into an open flow). Langflow loads all four at once. Or use the small button with an arrow pointing up, next to Projects in the left column. It says Upload a flow when you point at it.

  4. Give the flow a mail. Open the flow Triage (Enron). Copy this mail into the box Text of the component Text Input:

    From: Thomas Gygax <thomas.gygax@enron.com>
    To: Vince J Kaminski
    Date: Wed, 09 May 2001 03:49:00 GMT
    Subject: Clewlow/Strickland, "Energy Derivatives"
    
    Dear Vince,
    
    I'm working in the Continental Power Structured Products group here in London, and we'd like to use the
    Clewlow/Strickland book "Energy Derivatives. Pricing & Risk Management".
    Is there such a thing as an "Enron discount", and if so, how should we purchase the book (two copies) to get it?
    
    Thanks,
    THOMAS.
    

    If you see a yellow note "Legacy" on Text Input or Text Output, ignore it. If Langflow says Flow needs review, click Dismiss All.

  5. Run it. On the component Text Output, click ▶ (Run component). Wait a few seconds. Then click the small icon next to Output Text (Inspect output). You see "action": "reply" and a short reason after "why". Thomas asks a question, so the AI assistant would draft an answer. At the end you see "checks_ok": true: the component Check Triage checked the answer before it left Langflow.

  6. Change the mail. Replace the whole text in Text Input with this mail. Run it again:

    From: Thomas Gygax <thomas.gygax@enron.com>
    To: Vince J Kaminski
    Date: Wed, 09 May 2001 03:49:00 GMT
    Subject: Meeting next Tuesday
    
    Dear Vince,
    
    I'm working in the Continental Power Structured Products group here in London. I will be in Houston next week.
    Could we meet on Tuesday at 10 am in your office?
    
    Thanks,
    THOMAS.
    

    Now the action is calendar. Thomas wants a meeting, so the AI assistant would mark the mail for the calendar.

    Done testing? You can now delete the mail in Text Input. On 9 November, n8n will put in each new mail itself.

Well done: an AI just sorted a real mail for you. This flow is the brain of your AI assistant. On 9 November, n8n will give it hands: n8n will notice new mail, then reach out to this Langflow flow, check the answer, and act in your test Gmail account.

Stuck? Help for step 8

Part C · Set up n8n and test it

Gmailyour test Gmail accountPart A · steps 1–7 ✓
Mailkingo-pod's service: copies the mails, masks addressesPart A · steps 1–7 ✓
PostgreSQLevery mail as a rowPart A · steps 1–7 ✓
n8ndoes the workPart C · steps 9–13 · you are here
Langflowthinks: sorts and draftsPart B · step 8 ✓
Part A · steps 1–7 ✓
Part C · steps 9–13 · you are here
Part B · step 8 ✓

n8n is the second tool of the course. In n8n you build workflows. A workflow is like a flow in Langflow: boxes (nodes) joined by lines. The difference: Langflow is good for building AI; n8n is good for doing things. An n8n workflow starts by itself (for example when a new mail arrives), passes data from box to box, asks Langflow, and acts, for example in Gmail.

We name our workflows WS2.0, WS2.1, and so on: WS2 stands for Workshop II, the number for the step of the course. WS2.0 is the first one; you import it in step 10. On 9 November, you will build more.

9. Open n8n and give it its logins: four credentials

Langflow and n8n each keep their own keys:

n8n needs four credentials. One of them lets n8n use your Langflow flows: for it, Langflow gives n8n a key of its own.

  1. Open n8n and make your account. Open http://localhost:5678. The first time, n8n asks you to make an account. It lives only on your laptop.

    • Set up owner account: Email (for example the address of your test Gmail account), First Name, Last Name, and a Password with 8 or more characters, at least 1 number and 1 capital letter. Write the password down: n8n asks for it when you sign in again. Click Next.
    • Customize n8n to you: answer the questions as you like, then click Get started.
    • Get paid features for free (forever): click Skip. The course does not need them.
    • n8n shows Sign in instead? Then you made an account before: sign in with it.
  2. Make a key in Langflow for n8n. Switch to the browser window of Langflow. There, go to the profile icon in the top right → Settings → Langflow API Keys → Add New (top right). In the window Create API Key:

    • Description: n8n, so you know later what the key is for.
    • If the window also shows an expiration date, leave it empty (No expiration). Then the key works for the whole course.

    Click Generate API Key. Copy the key and keep it somewhere you can find it again: Langflow shows it only once. You need it in step 9.3.

  3. Make the four credentials in n8n. Now you make four credentials, one after the other: one for kingo-pod's database, one for your Langflow flows, and two for OpenRouter. You make each one the same way:

    1. In n8n, click the + at the very top left, then New credential. A window opens: Add new credential, Select an app or service to connect to.
    2. Type the credential's type into the search box (for example Postgres), pick it, and click Continue.
    3. First give the credential its name. At the top left of the next window, n8n has put a name of its own, for example Postgres account or Header Auth account. Click that name, delete it, and type the name from the list below (Top left). Use exactly these names: the workflows you import later look for them by name.
    4. Fill in the fields and click Save (top right). Then close the window.
    • Credential 1 of 4: give n8n access to kingo-pod's database
      • Type: Postgres
      • Top left: Mail (n8n_workshop_mail_n8n)
      • Host: postgres
      • Database: classroom
      • User: n8n_workshop_mail_n8n
      • Password: kingo2026
      • Port: 5432
      • SSL: disable
    • Credential 2 of 4: give n8n access to your Langflow flows
      • Type: Header Auth
      • Top left: Langflow (Run API)
      • Name: x-api-key
      • Value: the Langflow key from step 9.2
    • Credential 3 of 4: give n8n access to OpenRouter, to make the mails searchable (WS2.0)
      • Type: Header Auth
      • Top left: OpenRouter (embeddings)
      • Name: Authorization
      • Value: type the word Bearer, then one space (press the space bar once), then paste your OpenRouter key. With your own key, it looks like this: Bearer sk-or-v1-0a1b2c3d…
    • Credential 4 of 4: give n8n access to OpenRouter's chat models, for the guard (9 November)
      • Type: OpenRouter
      • Top left: OpenRouter (chat)
      • API Key: your OpenRouter key
    Why two for OpenRouter?

    It is the same key twice. WS2.0 and the guard talk to OpenRouter in two different ways, and each way needs the key in its own form.

    Check: in the column on the left of n8n, click Overview (or Personal), then the tab Credentials (in the middle of the page). You see your four credentials:

    The tab Credentials in n8n with the four credentials

More help: n8n: Create a credential

Stuck? Help for step 9

10. Make Kaminski's mails searchable for your AI assistant: import WS2.0

  1. In n8n, click Create workflow. In the new workflow, open the menu … at the top, next to the workflow's name:

    The menu … at the top of n8n, next to the workflow's name

    In that menu, click Import → From file, and pick ws2_hw/n8n/ws2_0_embed_mails.json (in the folder from step 8.2).

  2. If a node shows a credential warning, open it and pick the credential with the same name. For example: double-click the node fetch_embeddings_bge_m3. Under Header Auth, click the field and pick OpenRouter (embeddings):

    The node fetch_embeddings_bge_m3 in n8n: under Header Auth, pick OpenRouter (embeddings)

    Then close the node with the ✕ at the top right.

  3. Turn the workflow on: click Publish at the top. In the window that opens, click Publish again. Publish means: the workflow now runs by itself, every minute, until you unpublish it.

  4. Keep its list of runs short: open the menu … again (as in step 10.1) → Settings. Set Save successful production executions to Do not save, then click Save. n8n now keeps only the runs that fail.

What WS2.0 does

It does not search: it keeps the search up to date. This is a RAG system, like one with ChromaDB. It stores every mail as an embedding: a long list of numbers that stands for what the mail is about. But this time the store lives in kingo-pod's PostgreSQL database, in the table chunks. Every minute, WS2.0 takes up to 320 mails that are not in it yet, gets their embeddings from an AI model on OpenRouter (bge-m3) and saves them there. So the search knows a new mail soon after it arrives. The first time, all 3,244 mails take about 12 minutes once step 7 is done.

Why PostgreSQL, not ChromaDB?

In Yanbo's workshop, the embeddings lived in ChromaDB. Here they live in PostgreSQL, next to the mails:

  • One place. The mails are already rows in PostgreSQL. Their embeddings sit right next to them, in the table chunks. Nothing is copied into a second store, so nothing gets out of step.
  • Meaning and words in one search. One query finds mails by meaning and by words, and mixes both lists.
  • Rules in the database. When your AI assistant drafts a reply, Mail Search sees only mails up to the date of the mail, and never a confidential one. The database itself makes sure of that, not the AI.
  • Both tools reach it. n8n writes the embeddings (WS2.0), Langflow reads them (Mail Search).
Who searches: Langflow

When your AI assistant needs facts, the component Mail Search (in the Langflow flows Ask Mailbox and Draft Reply) searches in two ways at once:

  • By meaning, as in any RAG system: it turns the question into numbers in the same way and finds the mails whose numbers are closest.
  • By words: embeddings are bad at finding exact words, such as a name, a number or a rare term. PostgreSQL has an index of words, like the index at the back of a book. With it, Mail Search also finds the mails that contain the question's words, fast.

Step 13 tests it.

Stuck? Help for step 10

11. Check that all mails arrived

When the upload of step 7 is done (about 30 minutes), run this in the terminal, in your kingo-pod folder:

./kingo mail status

You should see:

A line Last error that ends in mbsync took longer than 600 s is fine: kingo-pod uploads in rounds of 10 minutes, and that line only says a round was full.

In your test Gmail account, look at two folders. Gmail counts conversations, not mails (step 7), so its numbers are smaller:

Moving forward, you might notice that the addresses differ

In Gmail, they end in .example (for example vince.kaminski@enron.example): no mail can reach such an address. In kingo-pod's PostgreSQL database on your computer, and so in n8n and Langflow, they end in .com or .edu, as in Kaminski's real mails. In between, kingo-pod's service called Mail converts the addresses, both ways:

  • From Gmail to PostgreSQL, it turns .example into the real address. So your AI assistant sees the mails as they really were.
  • From PostgreSQL to Gmail, when your AI assistant sends a mail, it never sends to a real address of the case: a mail to one of Kaminski's contacts lands in your own test Gmail account instead.

So no mail goes out through Gmail to a real person by mistake.

Stuck? Help for step 11

12. Check that new mail arrives: a test mail from your SKKU account

From your SKKU mail account, send this mail to your test Gmail account. Copy the subject and the text, then sign with your first name:

Kimchi for Monday
Hi,

This weekend my family makes kimchi for the winter (gimjang).
I will bring a box of it for everyone at the workshop on Monday.

Within a minute, it is in the inbox of your test Gmail account. kingo-pod also copies it into its database: ./kingo mail status now shows one mail more.

What just happened

You sent the mail to Gmail. kingo-pod's service called Mail copied it from Gmail onto your laptop and put it into the PostgreSQL database. There you can look at it easily, for example in CloudBeaver (step 14). A minute later, WS2.0 even adds it to your RAG system (step 10), so Mail Search can find your new mail by what it says. You try that in step 13.

It works: the mail from your SKKU account just arrived in kingo-pod on your laptop.

Stuck? Help for step 12

13. Test the mail search: ask the mailbox

We also prepared a Langflow component for you: Mail Search. It finds the mails that fit a question, among all 3,244 mails. On 9 November, your AI assistant will use it to find facts for its drafts. The flow Ask Mailbox (Enron) uses it, so you test Mail Search through this flow.

  1. In Langflow, open the flow Ask Mailbox (Enron) → Playground, as usual. If Langflow says Flow needs review, click Dismiss All (as in step 8.4).
  2. Ask: What is Steve Leppard's email address? The answer gives steven.leppard@… and names the mail it comes from. Here the name Leppard helps: Mail Search finds the mail by its words.
  3. Now find your own mail from step 12. Ask: Who is sharing homemade Korean food? None of these words is in your mail, so this time Mail Search finds it by meaning (step 10). The answer tells you about your kimchi.
  4. Ask three questions of your own about Kaminski's work in spring 2001.

That's it: your laptop is ready for 9 November. There you will build your first n8n workflow, WS2.1, and then your AI assistant.

Stuck? Help for step 13

Optional · Look into the database

14. Look into the database with CloudBeaver

This part is optional: skip it if you like. CloudBeaver shows kingo-pod's database in your browser: http://localhost:8978. In Workshop II, it shows you what your AI assistant did, row by row.

First time in CloudBeaver? Open Using CloudBeaver first and do its steps 1 and 2. This page is not just extra help: without it, you cannot get in. CloudBeaver has two logins, one for CloudBeaver and one for the database, with different passwords. The page shows both. You do it only once.

  1. On the left, open Classroom (PostgreSQL) → Databases → classroom → Schemas → n8n_workshop_mail → Tables.
  2. Double-click the table messages, then click the tab Data.
  3. One row is one mail of Kaminski's mailbox. Scroll to the right: who sent it (from_addr), to whom (to_addrs), the subject (subject), the text (body_text).
  4. Find your test mail from step 12: sort the table by the column sent_at, newest first. To sort, move the mouse over the head of the column sent_at: small arrows appear. Click them until the newest mail is at the top. Your mail is the first row.

On 9 November, your n8n workflows work with exactly these rows.

Stuck? Help for step 14

If something goes wrong

Find your step. Still stuck? Write to Frank: then we fix it together.

Step 1

What you see What to do
The line stops with fatal: not a git repository Your kingo-pod folder came from the old ZIP download. Run ./kingo update, then ./kingo mode abp.
The line stops with another error Send Frank a screenshot of the error and of what git status shows (run it in your kingo-pod folder).

Step 5

What you see What to do
No app passwords page Turn on 2-Step Verification with text messages or the Google prompt. A security key alone is not enough.

Step 6

What you see What to do
Sign-in failed Paste the app password again (not your Google password). Check that 2-Step Verification is on.
the mail service is off Run ./kingo mode abp, then step 6 again.
You want to start again with another account Run ./kingo mail reset, then step 6 again.
kingo-pod asks only three questions, and not Upload them now? Your kingo-pod is older. Do step 1 again (the update). Then go on with step 7 and run ./kingo mail seed there.

Step 7

What you see What to do
already seeded Your account already has the mails. Look in Sent and All Mail.
cannot read the mails in your test Gmail account kingo-pod could not look into your account, so it uploaded nothing. Check your internet connection. Then run ./kingo mail seed.
The numbers in Gmail stop growing Be patient: kingo-pod uploads in rounds, and Gmail's numbers can stand still for up to 10 minutes. Reload Gmail. Keep your laptop awake. Still no change after 20 minutes? Run ./kingo update, then ./kingo mail seed again: it copies nothing twice and gives the upload a push. Still stuck? Send Frank everything that ./kingo mail status shows.
Gmail shows about 1,600 in All Mail, not 3,244 That is right: Gmail counts conversations, not mails. ./kingo mail status counts the mails.
./kingo mail status shows an error Copy the error into a mail to Frank.

Step 8

What you see What to do
Triage shows an error Check the name OPEN_ROUTER_API_KEY in Langflow. Check that your key still has credit.
A component says the API key is missing, or its API key field is empty You loaded the flows before you made OPEN_ROUTER_API_KEY. Langflow then empties the key fields. Delete the four flows and load them again (step 8.3).
OpenRouter and Check Triage are not connected At the bottom right of OpenRouter, set the menu to Model Response. Then drag a line from its dot to Answer on Check Triage.

Step 9

What you see What to do
The Postgres credential fails Host postgres (not localhost), SSL disable, the password kingo2026.

Step 10

What you see What to do
WS2.0 fails at fetch_embeddings_bge_m3 The node must use the credential OpenRouter (embeddings): see step 10.2. In that credential, the value must be the word Bearer, one space, then your key.

Step 11

What you see What to do
Gmail shows about 1,600 in All Mail, not 3,244 That is right: Gmail counts conversations, not mails. ./kingo mail status counts the mails.
./kingo mail status shows an error Copy the error into a mail to Frank.

Step 12

What you see What to do
The mail from your SKKU account does not arrive Wait a minute. Check the address. Look into the inbox of your test Gmail account.

Step 13

What you see What to do
Ask Mailbox does not find your kimchi mail WS2.0 adds a new mail a minute after it arrives. Wait a minute and ask again. Still nothing? Your flows may be old: see the next row.
Ask Mailbox shows Error building Component Agent with Token "NaN" is invalid Your flows are old. Download ws2_hw.zip again, delete the four flows in Langflow, and load the four new files (step 8.3).
Ask Mailbox gives no answer WS2.0 is not finished yet (it needs about 12 minutes after step 7), or OPEN_ROUTER_API_KEY is missing in Langflow.
A component says the API key is missing, or its API key field is empty You loaded the flows before you made OPEN_ROUTER_API_KEY. Langflow then empties the key fields. Delete the four flows and load them again (step 8.3).
OpenRouter and Agent are not connected At the bottom right of OpenRouter, set the menu to Language Model. Then drag a line from its dot to Language Model on Agent.

Step 14

What you see What to do
CloudBeaver shows No Connections, or Classroom (PostgreSQL) does not open You are not logged in yet. Do steps 1 and 2 of Using CloudBeaver: two logins, with two different passwords.
Agentic AI in Business · Workshop II · Frank Huettner · SKK GSB.