Repository navigation
docs(skills): ask once for a paid run, after the quote - #56
Merged
Merged
Conversation
Every run that spends credit already stops for the user's explicit yes at resolveConfirmation. So the skills no longer ask the user to approve the configuration before queuing. The agent makes its choices, queues the run, and asks once with the choices and the price together. The perturbation skill drops its yes on the local estimate. It stops early only when the estimate exceeds the balance. Bump to 1.0.5. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
TitouanCh
approved these changes
Oct 27, 2026
oriolpetithelical
approved these changes
Oct 27, 2026
dmiv-helical
added a commit
that referenced
this pull request
Oct 28, 2026
docs(skills): ask once for a paid run, after the quote (#56)
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
A paid run could need two approvals. The agent first asked the user to approve its guessed settings. Then it queued the run and asked again for the price.
Every run that spends credit already stops for an explicit yes at
resolveConfirmation. Queuing is free and can be rejected. So the first approval adds nothing.What
fine-tune-modelandcompute-embeddingsnow say this rule outright. The agent makes its choices, queues the run, and asks once with the choices and the price together.fine-tune-modelpicks the label column itself and names it in the request. It asks before queuing only when it cannot choose.fine-tune-modelnow also says an earlier "sounds good" is not approval for the run.in-silico-perturbationdrops its yes on the local estimate. It stops early only when the estimate exceeds the balance fromwhoami(). Each model still needs its own yes.plugin.jsonfiles.Unchanged
Explicit yes before every run, one approval per run, the label column check, and reject then queue again to change a setting.
Tests
43 passed locally.
🤖 Generated with Claude Code