Record the contracts-free comparison - #977
Merged
Merged
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What this changes
The six benchmark behaviours were translated to contracts-free Starlark and run once with the same gpt-5.6-luna model and no tools. All 6/6 passed, while the five Deed MCP runs passed 5/6 each.
The exact generated answers, hidden behaviour cases, sandbox choice and limitations are recorded in
benchmarks/STARLARK.md. RESULTS and ROADMAP now link the result and close the comparison-arm item.Why this way
Starlark is familiar to the model but executes without imports, files, network or processes. The scorer was tested both ways: references scored 6/6, and the known largest/fallback bug scored 5/6.
The result is recorded against the project, not explained away: these six tasks do not establish that contracts helped. A next benchmark must make contract value observable.
What it still does not do
One Starlark run is not an effect size. It also does not supply the second model family still requested by Stage 1.
Checks
AI assistance: GitHub Copilot implemented and verified this change.