The design language that makes your AI harness better at design.
JavaScript
Bump skill-behavior test lineup: gemini-3.7-flash (#598)
Paul Bakaus36457e191ffb
diff --git a/tests/skill-behavior/README.md b/tests/skill-behavior/README.mdindex 01054e417..c9bb5038a 100644--- a/tests/skill-behavior/README.md+++ b/tests/skill-behavior/README.md@@ -90,6 +90,12 @@ Against the current default lineup, two cells are the known floor:`redesign replaces DESIGN` is flaky on every model, and `critique closes` isflaky on gemini-3.6-flash. A regression is a failure beyond those two.+The Google slot in `DEFAULT_MODELS` moved to `gemini-3.7-flash` on 2026-08-15.+Every Gemini cell in the tables below was measured on 3.6-flash (or 3.5-flash+where marked), and per the cross-version rule further down, those results are+unmeasured on 3.7, not inherited. Re-run the sweep on the next Setup or routing+change and update the tables to the new column.+**Read any failure against the clock before calling it behavior.** The suite ranat a 300s per-test timeout until 2026-08-13, and for the workflow-contractscenarios that cap was below the runtime of a correct run. `initialized naturaldiff --git a/tests/skill-behavior/providers.mjs b/tests/skill-behavior/providers.mjsindex d6bf0dc8b..9d31e4acf 100644--- a/tests/skill-behavior/providers.mjs+++ b/tests/skill-behavior/providers.mjs@@ -127,7 +127,7 @@ export function getProviderOptions(modelId) {* its own floor:* IMPECCABLE_SKILL_BEHAVIOR_MODELS=gpt-5.6-luna,deepseek-v4-flash*/-export const DEFAULT_MODELS = ['claude-sonnet-5', 'gpt-5.6-terra', 'gemini-3.6-flash'];+export const DEFAULT_MODELS = ['claude-sonnet-5', 'gpt-5.6-terra', 'gemini-3.7-flash'];export function resolveModelList() {const override = process.env.IMPECCABLE_SKILL_BEHAVIOR_MODELS;