From 45b5c3409c4e512aea4cf0726e60c1f8d7ca7868 Mon Sep 17 00:00:00 2001 From: prasanna8585 <65734642+prasanna8585@users.noreply.github.com> Date: Fri, 28 Aug 2026 19:01:43 +0530 Subject: [PATCH] fix: fence relayed agent content so it cannot pose as instructions convertForeignEvent presents another agent's turn to the current agent as a plain-text "For context:" preamble followed by unframed relayed content -- the same channel the real user speaks on. The relayed text is attacker-reachable: whoever talks to the other agent steers what it says, and its tool results carry whatever the tool read (a webpage, an email, an API response). Without any framing, that content is indistinguishable from a genuine instruction to the receiving agent's model. Port adk-python's already-merged fencing mitigation for the identical code path (_present_other_agent_message / _fencing.py, google/adk-python@9ffe8be6): each relayed text payload is wrapped in explicit BEGIN/END markers, and the leading preamble states plainly that fenced content is data to read, never instructions to follow. Markers appearing inside a payload are elided first, so a payload cannot forge the end of its own fence and continue speaking as the framework -- this is the exact attack adk-python's own fix names explicitly in its test comments. Applies to relayed text, tool-call arguments, and tool-response results; tool names are elided but left unfenced since they read as part of the sentence. Sibling of the identical fix already applied to adk-go and adk-js. Adds FencingTest.java, a dedicated unit test file for Fencing.java (which previously had none -- its behavior was only exercised indirectly through ContentsTest fixtures that never contain a fence marker of their own). Covers elideQuoteMarkers directly (no markers, begin-only, end-only, a forged complete fence, repeated markers) and quoteUntrusted end to end, confirming a forged end marker in relayed content cannot close the real fence early and a forged begin marker cannot open a fake one. Updates ContentsTest's processRequest_noInvocationBranch_ includesBranchedEvent, which landed on main after this branch's original base and still asserted the pre-fencing unquoted preamble shape, to the fenced shape used by every other test in the file. --- .../google/adk/flows/llmflows/Contents.java | 35 ++-- .../google/adk/flows/llmflows/Fencing.java | 68 ++++++++ .../adk/flows/llmflows/ContentsTest.java | 36 +++-- .../adk/flows/llmflows/FencingTest.java | 150 ++++++++++++++++++ 4 files changed, 267 insertions(+), 22 deletions(-) create mode 100644 core/src/main/java/com/google/adk/flows/llmflows/Fencing.java create mode 100644 core/src/test/java/com/google/adk/flows/llmflows/FencingTest.java diff --git a/core/src/main/java/com/google/adk/flows/llmflows/Contents.java b/core/src/main/java/com/google/adk/flows/llmflows/Contents.java index f0bfcd09a..a772accb4 100644 --- a/core/src/main/java/com/google/adk/flows/llmflows/Contents.java +++ b/core/src/main/java/com/google/adk/flows/llmflows/Contents.java @@ -410,8 +410,13 @@ private static boolean isOtherAgentReply(String agentName, Event event) { /** * Converts an {@code event} authored by another agent to a 'contextual-only' event. * - *
Returns {@code null} when nothing but the "For context:" preamble survives the conversion, - * so the caller drops the event instead of sending a preamble with no context after it. + *
Returns {@code null} when nothing but the preamble survives the conversion, so the caller + * drops the event instead of sending a preamble with no context after it. + * + *
The relayed text is attacker-reachable: whoever talks to the other agent steers what it
+ * says, and its tool results carry whatever the tool read. Each relayed text payload is therefore
+ * fenced (see {@link Fencing}), and the leading part states that fenced content is data, so a
+ * payload has to be believed rather than merely obeyed.
*/
private static @Nullable Event convertForeignEvent(Event event) {
if (event.content().isEmpty()
@@ -421,7 +426,7 @@ private static boolean isOtherAgentReply(String agentName, Event event) {
}
List Some of what a request carries is attacker-reachable: another agent's turn, a tool result,
+ * anything a model was talked into emitting. It travels on the same text channel the real user
+ * speaks on, so text posing as a directive is otherwise indistinguishable from one.
+ *
+ * Fencing marks where such a payload starts and ends and says, in the message itself, that what
+ * sits between the markers is data to read and not instructions to follow. This raises the bar
+ * rather than closing the class: a model can still be talked round by text it was told to distrust.
+ * What it removes is the structural ambiguity.
+ *
+ * Ported from adk-python's flows/llm_flows/_fencing.py.
+ */
+final class Fencing {
+
+ static final String QUOTED_CONTENT_BEGIN = "<< Markers inside the text are elided first, so quoted content cannot forge the end of its own
+ * block and carry on speaking as the framework.
+ */
+ static String quoteUntrusted(String text) {
+ return QUOTED_CONTENT_BEGIN + "\n" + elideQuoteMarkers(text) + "\n" + QUOTED_CONTENT_END;
+ }
+}
diff --git a/core/src/test/java/com/google/adk/flows/llmflows/ContentsTest.java b/core/src/test/java/com/google/adk/flows/llmflows/ContentsTest.java
index ce7655333..5597ce8f2 100644
--- a/core/src/test/java/com/google/adk/flows/llmflows/ContentsTest.java
+++ b/core/src/test/java/com/google/adk/flows/llmflows/ContentsTest.java
@@ -427,14 +427,15 @@ public void convertForeignEvent_eventsFromOtherAgents_returnsContextualOnlyEvent
.containsExactly(
u1.content().get(),
Content.fromParts(
- Part.fromText("For context:"),
- Part.fromText("[other_agent] said: Some text"),
- Part.fromText(
- "[other_agent] called tool `tool1` with parameters: "
- + "{\"arg1\":\"value\",\"arg2\":[1,2]}")),
+ otherAgentPreamblePart(),
+ otherAgentPart("[other_agent] said:", "Some text"),
+ otherAgentPart(
+ "[other_agent] called tool `tool1` with parameters:",
+ "{\"arg1\":\"value\",\"arg2\":[1,2]}")),
Content.fromParts(
- Part.fromText("For context:"),
- Part.fromText("[other_agent] `tool1` tool returned result: {\"result\":\"ok\"}")),
+ otherAgentPreamblePart(),
+ otherAgentPart(
+ "[other_agent] `tool1` tool returned result:", "{\"result\":\"ok\"}")),
a1.content().get(),
fr2.content().get())
.inOrder();
@@ -464,8 +465,8 @@ public void processRequest_includeContentsNone_lastEventIsOtherAgent() {
assertThat(result)
.containsExactly(
Content.fromParts(
- Part.fromText("For context:"),
- Part.fromText("[other_agent] said: Other Agent Turn")));
+ otherAgentPreamblePart(),
+ otherAgentPart("[other_agent] said:", "Other Agent Turn")));
}
@Test
@@ -1201,7 +1202,9 @@ public void processRequest_thoughtTextFromOtherAgent_isNotNarrated() {
contents.get(1).parts().get().stream()
.map(part -> part.text().orElse(""))
.collect(toImmutableList()))
- .containsExactly("For context:", "[" + OTHER_AGENT + "] said: It is in Paris.");
+ .containsExactly(
+ Fencing.OTHER_AGENT_CONTEXT_PREAMBLE,
+ "[" + OTHER_AGENT + "] said:\n" + Fencing.quoteUntrusted("It is in Paris."));
}
// The other-agent path still narrates what it can: media parts pass through unchanged, so the
@@ -1234,7 +1237,8 @@ public void processRequest_mediaPartFromOtherAgent_isKept() {
assertThat(contents).hasSize(2);
assertThat(contents.get(1).parts().get()).hasSize(2);
- assertThat(contents.get(1).parts().get().get(0).text()).hasValue("For context:");
+ assertThat(contents.get(1).parts().get().get(0).text())
+ .hasValue(Fencing.OTHER_AGENT_CONTEXT_PREAMBLE);
assertThat(contents.get(1).parts().get().get(1).inlineData()).isPresent();
}
@@ -1352,7 +1356,7 @@ public void processRequest_noInvocationBranch_includesBranchedEvent() {
assertThat(result)
.containsExactly(
Content.fromParts(
- Part.fromText("For context:"), Part.fromText("[agent_1] said: sibling output")));
+ otherAgentPreamblePart(), otherAgentPart("[agent_1] said:", "sibling output")));
}
private static Event createUserEvent(String id, String text) {
@@ -1364,6 +1368,14 @@ private static Event createUserEvent(String id, String text) {
.build();
}
+ private static Part otherAgentPreamblePart() {
+ return Part.fromText(Fencing.OTHER_AGENT_CONTEXT_PREAMBLE);
+ }
+
+ private static Part otherAgentPart(String attribution, String payload) {
+ return Part.fromText(attribution + "\n" + Fencing.quoteUntrusted(payload));
+ }
+
private static Event createUserEvent(
String id, String text, String invocationId, long timestamp) {
return Event.builder()
diff --git a/core/src/test/java/com/google/adk/flows/llmflows/FencingTest.java b/core/src/test/java/com/google/adk/flows/llmflows/FencingTest.java
new file mode 100644
index 000000000..fbfc60bd6
--- /dev/null
+++ b/core/src/test/java/com/google/adk/flows/llmflows/FencingTest.java
@@ -0,0 +1,150 @@
+/*
+ * Copyright 2026 Google LLC
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.google.adk.flows.llmflows;
+
+import static com.google.common.truth.Truth.assertThat;
+
+import org.junit.Test;
+import org.junit.runner.RunWith;
+import org.junit.runners.JUnit4;
+
+/**
+ * Unit tests for {@link Fencing}, in particular {@link Fencing#elideQuoteMarkers}, which the
+ * existing fixtures in {@link ContentsTest} exercise only indirectly through content that never
+ * contains a fence marker of its own. These tests cover the elision behavior directly: relayed
+ * content that contains a literal {@code <<