Rework the app-building skills from measured agent behaviour - #111
Open
onelxj wants to merge 2 commits into
Open
Rework the app-building skills from measured agent behaviour#111onelxj wants to merge 2 commits into
onelxj wants to merge 2 commits into
Conversation
Current Aviator status
This PR is not ready to merge (currently in state pending): this PR has not been approved. Pending Status Checks
See the real-time status of this PR on the
Aviator webapp.
Use the Aviator Chrome Extension
to see the status of your PR within GitHub.
|
onelxj
marked this pull request as draft
July 28, 2026 10:47
onelxj
force-pushed
the
onelxj.reboot-2
branch
3 times, most recently
from
July 28, 2026 12:30
41cde96 to
264d846
Compare
Six headless builds of the same app, with every token in the transcript attributed to what put it in the context. A third of the first run's cost was the agent reading Reboot's own source — authorizers.py, the pydantic compiler, idempotency.py, aborted.py, the codegen template — to answer questions these skills didn't. That is the most expensive way to learn anything: the request carries the whole context and adds more source to it for every later turn.
onelxj
commented
Jul 28, 2026
| [`references/react-client.md`](references/react-client.md). Read it | ||
| at the frontend step; don't reconstruct it from memory here. | ||
|
|
||
| ## Which References to Read, and When |
Contributor
Author
There was a problem hiding this comment.
I'll apply the same to chat-app as well.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
I've run the reboot agent on the same prompt a couple of times; the very first run (current main) cost $39 and took 45 min to build. Now we've removed the plan stage and started writing files early which produces still high-quality output (at least on the example I've used). The comparison was done against supabase project ($10 and ~16 min), but supabase didn't end up writing a backend servicer, only sql migration file + web, so arguably we are now on the same stage but with a much bigger skill-set.
With the skill changes I've been able to build the app in ~16 min and ~$12, which is a huge improvement already.