fix(evals): refine plan mode eval prompt to ensure toolchain completion

Align the 'foo' test prompt with existing project conventions while ensuring the model has the 'informal agreement' signal required to proceed to formal approval and implementation.
This commit is contained in:
Mahima Shanware
2026-04-09 18:13:35 +00:00
parent 08ae3ff9b8
commit 74ce3eef0c
+2 -2
View File
@@ -285,13 +285,13 @@ describe('plan_mode', () => {
},
});
evalTest('ALWAYS_PASSES', {
evalTest('USUALLY_PASSES', {
name: 'should transition from plan mode to normal execution and create a plan file from scratch',
params: {
settings,
},
prompt:
'Enter plan mode and plan to create a new module called foo. The plan should be saved as foo-plan.md. Then, exit plan mode.',
'Enter plan mode and plan to create a new module called foo. The plan should be saved as foo-plan.md. I agree with the strategy, so please create a detailed implementation plan and then execute it.',
assert: async (rig, result) => {
const enterPlanCalled = await rig.waitForToolCall('enter_plan_mode');
expect(