fix: address review on cleared test history and zero-count memory

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

backend/windmill-worker/src/ai_executor.rs

@@ -654,7 +654,9 @@ pub async fn handle_ai_agent_job(
         };
         // Only after interpolating the resource: these are caller-controlled and already resolved by
         // build_args_map, so passing them through it again would expand contextual values —
-        // `$WM_TOKEN` in a user message would reach the model provider.
+        // `$WM_TOKEN` in a user message would reach the model provider. The resource is not
+        // validated against a schema, so a flow-local key it happens to carry is dropped rather
+        // than read as the step's.
         for key in [
             "user_message",
             "user_attachments",
@@ -662,6 +664,7 @@ pub async fn handle_ai_agent_job(
             "memory_id",
             "previous_messages",
         ] {
+            brain.remove(key);
             if let Some(v) = local_args.get(key) {
                 brain.insert(
                     key.to_string(),

docs/reusable-ai-agents.md

@@ -68,7 +68,7 @@ The worker reconciles them once per agent invocation, nested agent tools include
 
 1. A legacy `auto` or `manual` memory: read as the editor that wrote it ran it. `manual` replays
    its list; `auto` uses the run's memory id, else the id baked into it, else runs stateless.
-   Neither history input is read.
+   Neither history input is read. An `auto` without a count, or with 0, is off and read as such.
 2. Managed memory: the memory id is the step's, else the run's. With no memory id the agent runs
    stateless, and a step `previous_messages` is ignored.
 3. Memory off: the history is `previous_messages`, else nothing. Memory is neither read nor

frontend/src/lib/components/ModuleTest.svelte

@@ -175,17 +175,19 @@
 			// it carries every brain key as undefined even though the form renders only the flow-local
 			// ones (`flowLocalAgentSchema`). Overlaying those would shadow the brain the draft just
 			// supplied with nothing, so an inlined step takes only the inputs its form actually offers.
-			// A history input left blank here stays as the step authored it: turned into an expression that
-			// evaluates to nothing, it would read as a memory id set to empty, where the step's own blank
-			// static value reads as unset.
+			// A history input left blank here is unset, as a blank static value is on the step: sent as an
+			// expression that evaluates to nothing it would read as a memory id set to empty, and left to
+			// the step's own transform it would reuse a value the author just cleared.
+			const isBlank = (v: unknown) => v == undefined || v === '' || (Array.isArray(v) && !v.length)
 			const formKeys = (
 				draft ? (AGENT_FLOW_LOCAL_KEYS as readonly string[]) : Object.keys(args)
 			).filter(
-				(key) =>
-					!(AGENT_HISTORY_KEYS as readonly string[]).includes(key) ||
-					(args[key] != undefined &&
-						args[key] !== '' &&
-						!(Array.isArray(args[key]) && !args[key].length))
+				(key) => !(AGENT_HISTORY_KEYS as readonly string[]).includes(key) || !isBlank(args[key])
+			)
+			const stepTransforms = Object.fromEntries(
+				Object.entries((agentVal.input_transforms ?? {}) as Record<string, InputTransform>).filter(
+					([key]) => !(AGENT_HISTORY_KEYS as readonly string[]).includes(key) || !isBlank(args[key])
+				)
 			)
 
 			// The test form only covers the schema it was given, and for a standalone agent that may be
@@ -194,9 +196,7 @@
 			// in the form after the test panel mounted is what runs. A linked agent needs none of this:
 			// the server reads its brain from the resource.
 			const inputTransforms: { [key: string]: JavascriptTransform | InputTransform } = {
-				...(agentVal.agent
-					? {}
-					: ((agentVal.input_transforms ?? {}) as Record<string, InputTransform>)),
+				...(agentVal.agent ? {} : stepTransforms),
 				...Object.fromEntries(
 					formKeys.map((key) => [
 						key,

frontend/src/lib/components/flows/agentFormFields.test.ts

@@ -111,6 +111,9 @@ describe('memoryOptionLabel', () => {
 		expect(memoryOptionLabel({ kind: 'manual', messages: [] })).toBe('Previous messages (legacy)')
 		expect(memoryOptionLabel({ kind: 'auto', context_length: 4 })).toBe('On (legacy)')
 		expect(memoryOptionLabel({ kind: 'window', context_length: 10 })).toBe('On')
+		// Keeping no messages runs as off, whichever kind says so.
+		expect(memoryOptionLabel({ kind: 'window', context_length: 0 })).toBe('Off')
+		expect(memoryOptionLabel({ kind: 'auto' })).toBe('Off')
 		expect(memoryOptionLabel(undefined)).toBeUndefined()
 	})
 })

frontend/src/lib/components/flows/flowInfers.ts

@@ -16,6 +16,11 @@ export const MEMORY_OPTION_LABELS: Record<string, string> = {
 }
 
 export function memoryOptionLabel(memory: any): string | undefined {
+	// Managed memory that keeps no messages runs as off, and a note about what that state reads
+	// must say so.
+	if ((memory?.kind === 'window' || memory?.kind === 'auto') && !memory.context_length) {
+		return MEMORY_OPTION_LABELS.off
+	}
 	return memory?.kind ? MEMORY_OPTION_LABELS[memory.kind] : undefined
 }
 
@@ -72,7 +77,7 @@ export const AI_AGENT_SCHEMA: Schema = {
 						context_length: {
 							type: 'number',
 							title: 'Messages to keep',
-							description: 'Number of most recent messages to load and store.',
+							description: 'Number of most recent messages to load and store. 0 turns memory off.',
 							default: 10
 						}
 					},

frontend/src/lib/components/flows/utils.svelte.ts

@@ -154,7 +154,8 @@ export function cleanFlow(flow: OpenFlow | any): OpenFlow & {
  * A chat flow runs with the conversation as its memory id, so an id an older editor baked into a
  * step's memory was never read there and is dropped. Anywhere else it still applies to runs that
  * pass none, and stays until the author converts it. An empty static memory id or message list
- * reads as unset at runtime, so neither is persisted.
+ * reads as unset at runtime, so neither is persisted, and managed memory that keeps no messages
+ * runs as off, so it is saved as off.
  */
 export function normalizeAgentHistory(
 	inputTransforms: Record<string, any> | undefined,
@@ -162,6 +163,13 @@ export function normalizeAgentHistory(
 ) {
 	if (!inputTransforms) return
 	const memory = inputTransforms.memory
+	if (
+		memory?.type === 'static' &&
+		memory.value?.kind === 'window' &&
+		!memory.value.context_length
+	) {
+		memory.value = { kind: 'off' }
+	}
 	if (
 		memory?.type === 'static' &&
 		memory.value?.kind === 'auto' &&

frontend/src/lib/components/flows/utils.test.ts

@@ -78,6 +78,16 @@ describe('normalizeAgentHistory', () => {
 		expect(transforms.memory.value).toEqual({ kind: 'auto', context_length: 10 })
 	})
 
+	it('saves managed memory that keeps no messages as off, which is how it runs', () => {
+		for (const context_length of [0, null, undefined]) {
+			const transforms: Record<string, any> = {
+				memory: { type: 'static', value: { kind: 'window', context_length } }
+			}
+			normalizeAgentHistory(transforms, false)
+			expect(transforms.memory.value).toEqual({ kind: 'off' })
+		}
+	})
+
 	it('does not persist an empty static memory id or message list', () => {
 		const transforms: Record<string, any> = {
 			memory_id: { type: 'static', value: ' ' },

openflow.openapi.yaml

@@ -556,7 +556,7 @@ components:
       description: |
         Deprecated, still read as it was written: the run's memory id, else the `memory_id` here.
         The step's own `memory_id` is not read while this kind is set; switch the kind to `window`
-        to use it.
+        to use it. Without a `context_length`, or with 0, it is `off` and reads `previous_messages`.
       properties:
         kind:
           type: string