First run: the first skill step opens the catalog - #180
Conversation
The step sent people to their own library, which on day zero is empty. It now opens /discover, where saving a skill is one click, and keeps the GitHub repository path as a secondary button. The event records which destination was taken.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Team Run ID: 📒 Files selected for processing (3)
Included review availability: Your plan provides up to 1 included review per hour; 0 remain after this review. WalkthroughThe first-skill onboarding step now opens ChangesCatalog-first onboarding flow
Estimated code review effort: 2 (Simple) | ~10 minutes Merge Risk: ⚪ Minimal · up to The first-skill onboarding card now prioritizes the catalog while retaining repository import, with destination-specific analytics and coverage for both paths. No concrete current-head merge risk remains. Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
Full details: Description checkExplanation The description gives detailed context, rationale, metrics, and validation results. However, it does not follow the repository template: it uses "What changes" instead of "What changed," omits the required "Validation" and "Impact" sections, and does not provide the required impact checklist information. Resolution Rename "What changes" to "What changed" and add the required "Validation" and "Impact" sections. Include the validation checklist, document that no database or environment changes exist, confirm that no secrets or private skill content are included, and confirm that saved-skill ownership is unchanged.
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
@CodeRabbit review |
✅ Action performedReview finished.
|
What changes
On
/start, the "Add your first skill" card now leads to the public catalog./discover./library.onboarding_step_clickedgains an optionaldestinationproperty (discoverorlibrary), set only by the first-skill step. Older events have no destination and meant/library.Why
Since 23 August we created 7 teams. 5 of them saw
/start. 2 clicked the first-skill step, which sent them to/library: an empty library whose empty state then offers "Add a skill" (a GitHub repository dialog) or "Find skills" (a link to/discover). One of those two teams walked that extra hop, reached/discoverand saved 3 skills in 4 minutes. The other landed on the empty page and left. 6 of the 7 teams are still at 0 skills.The only team that saved anything did it from the catalog, and the step that was meant to produce a first skill was pointing at the one page that has nothing in it yet.
Hypothesis
A new team has no repository URL ready, but it can recognize a useful skill in a catalog and save it in one click. Landing the first-skill step on
/discovershould turn more clicks on that step into a first saved skill. Keeping the repository path visible as a secondary button costs the people who do have a URL only one extra click.Metric that should move
onboarding_step_clickedwithstep=first_skillfollowed byskill_savedin the same session.Secondary, for reading the split: the
destinationproperty on the step event tells us how many people want the catalog versus the repository path.How we will know
Compare teams created after the merge with the 7 created before it. At 4 or 5 new teams a month, a readable difference will take weeks and the numbers will stay noisy the whole time. This is a directional bet on a broken step, not a measurable experiment; do not expect significance.
Checks
node --test tests/*.test.mjs: 586 tests, all passing.tsc --noEmit -p .: clean.Summary by CodeRabbit
New Features
Tests