diff --git a/AGENTS.md b/AGENTS.md
new file mode 100644
index 0000000..bc0803f
--- /dev/null
+++ b/AGENTS.md
@@ -0,0 +1,65 @@
+# AGENTS.md
+
+Behavioral guidelines to reduce common LLM coding mistakes. Merge with project-specific instructions as needed.
+
+**Tradeoff:** These guidelines bias toward caution over speed. For trivial tasks, use judgment.
+
+## 1. Think Before Coding
+
+**Don't assume. Don't hide confusion. Surface tradeoffs.**
+
+Before implementing:
+- State your assumptions explicitly. If uncertain, ask.
+- If multiple interpretations exist, present them - don't pick silently.
+- If a simpler approach exists, say so. Push back when warranted.
+- If something is unclear, stop. Name what's confusing. Ask.
+
+## 2. Simplicity First
+
+**Minimum code that solves the problem. Nothing speculative.**
+
+- No features beyond what was asked.
+- No abstractions for single-use code.
+- No "flexibility" or "configurability" that wasn't requested.
+- No error handling for impossible scenarios.
+- If you write 200 lines and it could be 50, rewrite it.
+
+Ask yourself: "Would a senior engineer say this is overcomplicated?" If yes, simplify.
+
+## 3. Surgical Changes
+
+**Touch only what you must. Clean up only your own mess.**
+
+When editing existing code:
+- Don't "improve" adjacent code, comments, or formatting.
+- Don't refactor things that aren't broken.
+- Match existing style, even if you'd do it differently.
+- If you notice unrelated dead code, mention it - don't delete it.
+
+When your changes create orphans:
+- Remove imports/variables/functions that YOUR changes made unused.
+- Don't remove pre-existing dead code unless asked.
+
+The test: Every changed line should trace directly to the user's request.
+
+## 4. Goal-Driven Execution
+
+**Define success criteria. Loop until verified.**
+
+Transform tasks into verifiable goals:
+- "Add validation" → "Write tests for invalid inputs, then make them pass"
+- "Fix the bug" → "Write a test that reproduces it, then make it pass"
+- "Refactor X" → "Ensure tests pass before and after"
+
+For multi-step tasks, state a brief plan:
+```
+1. [Step] → verify: [check]
+2. [Step] → verify: [check]
+3. [Step] → verify: [check]
+```
+
+Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.
+
+---
+
+**These guidelines are working if:** fewer unnecessary changes in diffs, fewer rewrites due to overcomplication, and clarifying questions come before implementation rather than after mistakes.
diff --git a/frontend/src/components/screener/StrategyBuilderDialog.tsx b/frontend/src/components/screener/StrategyBuilderDialog.tsx
index 850523a..39566c9 100644
--- a/frontend/src/components/screener/StrategyBuilderDialog.tsx
+++ b/frontend/src/components/screener/StrategyBuilderDialog.tsx
@@ -181,6 +181,7 @@ export function StrategyBuilderDialog({ open, onClose, onSavedId, mode = 'create
   useEffect(() => { if (loaded) persist() }, [loaded, persist])
 
   const clearDraft = () => {
+    draftStore.set(null)
     setName(''); setDescription(''); setDirection('long')
     setRules(''); setCode(''); setStep(1); setError(''); setInstruction('')
     setStrategyId(''); setSource('ai'); setValidated(false)
@@ -288,11 +289,11 @@ export function StrategyBuilderDialog({ open, onClose, onSavedId, mode = 'create
         name: name.trim(),
         description: description.trim(),
       })
+      clearDraft()
       const genRules = parseRules(draftCode)
       const finalRules = (genRules || rules).trim()
       if (finalRules) { const saved = storage.strategyRules.get({}); saved[id] = finalRules; storage.strategyRules.set(saved) }
       await onSavedId?.(id)
-      clearDraft()
       setTimeout(() => onClose(), 1000)
     } catch (e: any) { setError(String(e?.message ?? '保存失败')) }
     setSaving(false)
diff --git a/frontend/src/pages/Screener.tsx b/frontend/src/pages/Screener.tsx
index 39e3d7f..760117d 100644
--- a/frontend/src/pages/Screener.tsx
+++ b/frontend/src/pages/Screener.tsx
@@ -977,7 +977,7 @@ export function Screener() {
         onClose={() => setShowBuilder(false)}
         mode={builderMode}
         onSavedId={async id => {
-          const data = await qc.fetchQuery({ queryKey: QK.screenerStrategies('stock'), queryFn: () => api.screenerStrategies('stock') })
+          const data = await qc.fetchQuery({ queryKey: QK.screenerStrategies('stock'), queryFn: () => api.screenerStrategies('stock'), staleTime: 0 })
           if (!data.presets.some(s => s.id === id)) {
             throw new Error(`策略 ${id} 已保存但未加载，请检查策略代码`)
           }