feat: new onboarding flow by lifeizhou-ap · Pull Request #7266 · block/goose

lifeizhou-ap · 2026-02-17T08:34:11Z

Key changes:

Two paths for users to start to reduce decision fatigue. (old flow: 6 options in one page and with multiple recommend badges)
Contents expand inline below the cards to keep user stay in one place instead of using modal or navigating to new screen.
Having a success page with the privacy choice. In the old flow, the privacy choice is almost hidden at the end of the page so that user cannot see it.
Using the dropdown with auto completion for users to easily find the providers. Also in each provider list the important field for user to enter and provide the "get api key" instructions for major providers
Using provider default model to reduce frictions so that users can use goose quickly

Type of Change

Testing

Manual testing and unit test

New Experience videos

onboarding_nanogpt.mov

onboarding_local_model.mov

Screen.Recording.2026-03-16.at.2.36.16.pm.mov

… fields

* main: (73 commits) dev: add cmake to hermitized env (#7399) refactor: remove allows_unlisted_models flag, always allow custom model entry (#7255) feat: expose context window utilization to agent via MOIM (#7418) Small model naming (#7394) chore(deps): bump ajv in /documentation (#7416) doc: groq models (#7404) Client settings (#7381) Fix settings tabs getting cut off in narrow windows (#7379) docs: voice dictation updates (#7396) [docs] Add Excalidraw MCP App Tutorial (#7401) Post release checklist as a comment on release PRs (#7307) unique api key (#7391) fix: use correct colors for download progress bar (#7390) Add local model settings access from bottom bar model menu (#7378) Change Recipe Security Scanner API key (#7387) switch Ask AI Discord bot from openrouter to anthropic (#7386) feat(ui): show token counts directly for "free" providers (#7383) Update creator note (#7384) Remove display_name from local model API and use model ID everywhere (#7382) fix(summon): stop MOIM from telling models to sleep while waiting for tasks (#7377) ...

Copilot

Pull request overview

Copilot reviewed 34 out of 34 changed files in this pull request and generated 4 comments.

Copilot · 2026-02-24T06:45:18Z

ui/desktop/src/components/onboarding/FreeOptionCards.tsx

+  const handleSetup = async (type: typeof TETRATE | typeof NANOGPT) => {
+    setError(null);
+    setSelectedProvider(type);
+    try {
+      const result = type === TETRATE ? await startTetrateSetup() : await startNanogptSetup();
+      if (result.success) {
+        onConfigured(type);
+      } else {
+        setError({ message: result.message, type });
+      }
+    } catch {
+      setError({ message: 'An unexpected error occurred during setup.', type });
+    }
+  };


The analytics tracking for provider selection is missing for TETRATE and NANOGPT providers. Only LOCAL_PROVIDER calls trackOnboardingProviderSelected. Consider calling this function in the handleSetup function after a successful result to track all provider selections consistently.

Centralise send ing the event for all providers in handleConfigure function

Copilot · 2026-02-24T06:45:19Z

crates/goose/src/providers/nanogpt.rs

+            vec![NANOGPT_DEFAULT_MODEL],
+            NANOGPT_DOC_URL,
+            vec![ConfigKey::new(NANOGPT_API_KEY, true, true, None, true)],
+        )


The NanoGPT provider metadata doesn't include setup_steps, unlike other providers (OpenAI, Anthropic, Google, OpenRouter) that were updated in this PR to include them. Consider adding setup_steps with instructions for obtaining a NanoGPT API key for consistency, or explicitly documenting why it's omitted if the authentication flow is OAuth-based.

Suggested change

)

)

.with_setup_steps(vec![

"Create an account at NanoGPT (https://nano-gpt.com/) and sign in to the dashboard."

.into(),

"From the NanoGPT dashboard, generate an API key with access to the models you want to use."

.into(),

"Configure the generated key as NANOGPT_API_KEY in your goose configuration or environment secrets."

.into(),

])

we can add this later

Copilot · 2026-02-24T06:45:19Z

ui/desktop/src/api/types.gen.ts

+export type CreateCustomProviderResponse2 = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];



Type naming collision: CreateCustomProviderResponse is defined both as the schema type at line 130 and as a union type at line 2170 (with CreateCustomProviderResponse2 being the actual union type). This creates confusion. The generated code should use consistent naming to avoid this duplication.

Suggested change

export type CreateCustomProviderResponse2 = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];

export type CreateCustomProviderResponseUnion = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];

/**

* @deprecated Use CreateCustomProviderResponseUnion instead.

*/

export type CreateCustomProviderResponse2 = CreateCustomProviderResponseUnion;

this seems fine, the naming method is weird but this is just how the sdk generator works

Copilot · 2026-02-24T06:45:20Z

ui/desktop/src/components/onboarding/ProviderSelector.tsx

+  const handleCreateCustomProvider = async (data: UpdateCustomProviderRequest) => {
+    const result = await createCustomProvider({ body: data, throwOnError: true });
+    setShowCustomModal(false);
+    if (result.data?.provider_name) {
+      onConfigured(result.data.provider_name);
+    }
+  };


Missing error handling for handleCreateCustomProvider. If the createCustomProvider call fails (despite throwOnError: true), the error will be unhandled and could crash the app. The catch block in CustomProviderForm won't handle this since it's called from a different component. Consider wrapping this in a try-catch block to handle API errors gracefully.

Changed the CustomProviderForm to handle the error, no need try catch here

* main: (191 commits) fix: add tool_choice and parallel_tool_calls to chatgpt_codex provider (#7867) fix: tool confirmation handling for multiple requests (#7856) Remove dead OllamaSetup onboarding flow (#7861) fix: resolve tokio::sync::Mutex deadlock in recipe retry path (#7832) Upgrade Electron 40.6.0 → 41.0.0 (#7851) Only show up to 50 lines of source code (#7578) fix: stop writing without error when hitting broken pipe for goose session list (#7858) feat(acp): add session/set_mode handler (#7801) Keep messages in sync (#7850) More acp tools (#7843) fix: skip upgrade-insecure-requests CSP for external HTTP backends (#7714) fix(shell): prevent hang when command backgrounds a child process (#7689) Remove include from Cargo.toml in goose-mcp (#7838) Exit agent loop when tool call JSON fails to parse (#7840) chore: remove redundant husky prepare script (#7829) Add github actions workflow for unused deps (#7681) fix: prevent SSE connection drops from silently truncating responses (#7831) doc: Added notes in contribution guide for pnpm (#7833) add prefer-offline to pnpm config to skip unnecessary registry lookups (#7828) fix: remove dead read handler from DeveloperClient (#7821) ...

* main: updated canonical models (#7920) feat(autovisualiser): Migrate the autovisualiser extension to MCP Apps (#7852)

jamadeo

A couple of nits but this looks nice!

I'm wondering also about the value of the feature flag. Are you thinking that we might want to do a build where we turn this off? Because there is always git revert

jamadeo · 2026-03-16T19:44:40Z

crates/goose/src/providers/anthropic.rs

+            "Go to https://console.anthropic.com and sign up or log in",
+            "Click 'API Keys' in the left sidebar",
+            "Click 'Create Key'",
+            "Copy the key and paste it above",


or maybe

https://platform.claude.com/settings/keys

which is what anthropic docs seem to link to

jamadeo · 2026-03-16T19:46:06Z

crates/goose/src/providers/nanogpt.rs

+
+        let is_subscription = Self::check_subscription(&api_key).await;
+        let host = if is_subscription {
+            tracing::info!("NanoGPT subscription active, using subscription endpoint");


you can probably drop these traces

changed to debug

jamadeo · 2026-03-16T19:50:42Z

crates/goose/src/posthog.rs

+    // check so we can track the funnel before the user makes their choice.
+    let is_onboarding_event =
+        event_name.starts_with("onboarding_") || event_name == "telemetry_preference_set";
+    if !is_onboarding_event {


ideally we'd hold onto the event and then send it (or not) only if the user opts in -- think that's possible?

yes, I think we could. I was thinking about this too originally although the code would be a bit complicated. However, we would like to get the metrics for the onboarding as much as we can as requested by UX.

jamadeo · 2026-03-16T19:56:20Z

ui/desktop/src/api/types.gen.ts

+export type CreateCustomProviderResponse2 = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];



this seems fine, the naming method is weird but this is just how the sdk generator works

jamadeo · 2026-03-16T19:57:33Z

ui/desktop/src/components/onboarding/LocalModelPicker.tsx

+    const load = async () => {
+      try {
+        const response = await listLocalModels();
+        if (response.data) {
+          setModels(response.data);
+
+          const alreadyDownloaded = response.data.find((m) => m.status.state === 'Downloaded');
+          if (alreadyDownloaded) {
+            setSelectedModelId(alreadyDownloaded.id);
+          } else {
+            const recommended = response.data.find((m: LocalModelResponse) => m.recommended);
+            if (recommended) setSelectedModelId(recommended.id);
+          }
+        }
+      } catch (error) {
+        console.error('Failed to load local models:', error);
+        setErrorMessage('Failed to load available models. Please try again.');
+        setPhase('error');
+        return;
+      }
+      setPhase('select');
+    };
+    load();
+  }, []);


I think this is still valid?

jamadeo · 2026-03-16T20:07:13Z

ui/desktop/src/components/onboarding/ProviderSelector.tsx

+        >
+          <Key size={20} className="text-text-muted mb-2" />
+          <span className="font-medium text-text-default text-base block">
+            Use Your Own Provider


IMO the language here could be a little clearer: I don't think of Anthropic/OpenAI as "my own" provider. Maybe something like

| Use a Free/Local Provider | Connect to a Provider | | Use a local model or a service with free credits | Use your own login/key with OpenAI, Anthropic, Google, etc |

jamadeo · 2026-03-16T20:10:25Z

ui/desktop/src/components/onboarding/OnboardingSuccess.tsx

+              <h2 className="text-xl font-light text-text-default mb-1">
+                {providerName === LOCAL_PROVIDER
+                  ? 'Local model ready'
+                  : `Connected to ${providerName}`}


nit: providerName here seems to be the key, not the "display name" -- when I went through it said Connected to anthropic while I'd expect Anthropic (capital A)

jamadeo · 2026-03-16T20:14:04Z

ui/desktop/src/featureFlags.ts

@@ -0,0 +1 @@
+export const USE_NEW_ONBOARDING = true;


is this here in case we want to do a build where it is disabled?

this will be only changed in case the new onboarding flow does not work

lifeizhou-ap · 2026-03-16T23:35:52Z

A couple of nits but this looks nice!

I'm wondering also about the value of the feature flag. Are you thinking that we might want to do a build where we turn this off? Because there is always git revert

Hi @jamadeo,

Thanks for the review!

I added the feature toggle as this PR has lots of changes and also onboarding is a bit critical as this is the first experience after users download the app. In case it went wrong, we can change the feature toggle and launch a patch release. I'll remove it after one release and clean it up

* main: Add DCO git commit command to AGENTS.md (#7945) fix(claude-code): remove incorrect agent_visible filter on user message (#7931) No Check do Check (#7942) Log 500 errors and also show error for direct download (#7936) fix: retry on authentication failure with credential refresh (#7812) Remove java/.ai-usage-marker directory (#7925) test(acp): add terminal delegation fixtures and fix shell singleton (#7923) fix: bump pctx_code_mode to 0.3.0 for iterator type checking fix (#7892) feat: persist GooseMode per-session via session DB (#7854) feat(otel): propagate session.id to spans and log records (#7490) fix(test): add env_lock to is_openai_reasoning_model tests (#7917) fix(acp): pass session_id when loading extensions so skills are discovered (#7868)

* origin/main: feat: adversarial agent for preventing leaking of info and more (#7948) Update contributing.md (#7927) docs: add credit balance monitoring section (#7952) docs: add Cerebras provider to supported providers list (#7953) docs: add TUI client documentation to ACP clients guide (#7950) fix: removed double dash in pnpm command (#7951) docs: polish ACP docs (#7946) claude adaptive thinking (#7944) feat: new onboarding flow (#7266) Add DCO git commit command to AGENTS.md (#7945) fix(claude-code): remove incorrect agent_visible filter on user message (#7931) No Check do Check (#7942) Log 500 errors and also show error for direct download (#7936) fix: retry on authentication failure with credential refresh (#7812) Remove java/.ai-usage-marker directory (#7925) test(acp): add terminal delegation fixtures and fix shell singleton (#7923) fix: bump pctx_code_mode to 0.3.0 for iterator type checking fix (#7892) feat: persist GooseMode per-session via session DB (#7854)

Signed-off-by: esnyder <[email protected]>

lifeizhou-ap added 29 commits February 17, 2026 09:27

onboarding page to start

d0debc3

added primary attribute on configKey in provider

06dfcc9

cherry pick primary configkey change

75386e5

apply primary in cli config provider, also enables user to config all…

773d392

… fields

change SAGEMAKER_ENDPOINT_NAME as a required field

c99cbc9

fmt

f08a0b1

added some fallback

882a2b8

added props to show options in provider config component

122221e

onboarding page ui

6243dd0

merge main

992a64a

add instructions

a6e5e17

set default model for selected provider

831091b

crete feature toggle to switch between old/new onboarding

95c9361

add option to add custom provider

048059b

changed add custom provider response with provider name only

743d49e

enable nanogpt device login flow

46fa106

added nanogpt provider

535b6b5

text change

301d4f3

change font size

2e85a79

extract to constant

6f03f6f

clean up special case for ollama and better error message

7488025

added privacy and success page

2ce5e27

added tracking events

c504e2a

updated the instructions

affff89

error handling

d183cf0

feature toggle off by default

c1062da

added local model selection

bec2abf

fix test

898310f

lifeizhou-ap marked this pull request as ready for review February 23, 2026 06:48

error message and error handling

ae924c0

Copilot AI review requested due to automatic review settings February 24, 2026 06:41

Copilot started reviewing on behalf of lifeizhou-ap February 24, 2026 06:42 View session

Copilot AI reviewed Feb 24, 2026

View reviewed changes

track provider selected

a152100

lifeizhou-ap requested review from DOsinga, katzdave and zanesq February 24, 2026 09:31

lifeizhou-ap added 3 commits February 25, 2026 14:10

used nano-gpt to get canonical model data

4ec0e6f

update text position

5c959b6

fixed test

5454a49

DOsinga assigned jamadeo Mar 9, 2026

lifeizhou-ap added 5 commits March 16, 2026 12:44

updated local model style

45f83d1

tracking onboarding events

da52745

Merge branch 'main' into lifei/simplify_onboarding_flow

9cbe4d8

* main: updated canonical models (#7920) feat(autovisualiser): Migrate the autovisualiser extension to MCP Apps (#7852)

updated default model for nanogpt

636d007

jamadeo approved these changes Mar 16, 2026

View reviewed changes

lifeizhou-ap added 2 commits March 17, 2026 11:56

address review comments

51ef2dd

lifeizhou-ap added this pull request to the merge queue Mar 17, 2026

Merged via the queue into main with commit 4604354 Mar 17, 2026
29 of 30 checks passed

lifeizhou-ap deleted the lifei/simplify_onboarding_flow branch March 17, 2026 01:39

github-actions bot mentioned this pull request Mar 17, 2026

chore(release): release version 1.28.0 (minor) #7780

Open

elijahsgh pushed a commit to elijahsgh/goose that referenced this pull request Mar 21, 2026

feat: new onboarding flow (block#7266)

d1a4f1f

Signed-off-by: esnyder <[email protected]>

elijahsgh pushed a commit to elijahsgh/goose that referenced this pull request Mar 21, 2026

feat: new onboarding flow (block#7266)

291da88

Signed-off-by: esnyder <[email protected]>

lifeizhou-ap mentioned this pull request Mar 25, 2026

Lifei/clean up old onboarding flow #8099

Merged

-        )
+        )
+        .with_setup_steps(vec![
+            "Create an account at NanoGPT (https://nano-gpt.com/) and sign in to the dashboard."
+                .into(),
+            "From the NanoGPT dashboard, generate an API key with access to the models you want to use."
+                .into(),
+            "Configure the generated key as NANOGPT_API_KEY in your goose configuration or environment secrets."
+                .into(),
+        ])

		export type CreateCustomProviderResponse2 = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];

-export type CreateCustomProviderResponse2 = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];
+export type CreateCustomProviderResponseUnion = CreateCustomProviderResponses[keyof CreateCustomProviderResponses];
+/**
+ * @deprecated Use CreateCustomProviderResponseUnion instead.
+ */
+export type CreateCustomProviderResponse2 = CreateCustomProviderResponseUnion;

Conversation

lifeizhou-ap commented Feb 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Key changes:

Type of Change

Testing

New Experience videos

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jamadeo left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

lifeizhou-ap commented Mar 16, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

lifeizhou-ap commented Feb 17, 2026 •

edited

Loading