Repository navigation
feat: agentsync optimize suggests model and effort for agents - #19
Merged
Merged
Conversation
`agentsync optimize` asks a decision model (TypeSafe Jev by default, Cloudflare Clef optional) through OpenRouter which Claude model family and effort each agent should run on, and reports current against suggested with the advisor's confidence. It only suggests a change above --min-confidence, refuses to exceed --max-cost, and stops when spend can't be tracked. Report only; nothing is written. Needs $OPENROUTER_API_KEY; every other command stays offline.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
--min-confidence; a part below the threshold is marked low-confidence on its own, per part.--max-cost) is estimated up front and tracked from the reported cost; the run stops when spend can't be tracked.Verification
just checkpasses on macOS and Linux. New tests were confirmed to fail when the guarded behaviour is broken. Live runs against real agents with Jev and Clef cost under $0.002.Report only:
--applyfollows in a separate PR.Refs #2