fix: preserve project instruction wrappers

Co-authored-by: Codex noreply@openai.com

Author: viyatb-oai

codex-rs/instructions/src/user_instructions.rs

@@ -8,6 +8,8 @@ use crate::fragment::AGENTS_MD_START_MARKER;
 use crate::fragment::SKILL_FRAGMENT;
 
 pub const USER_INSTRUCTIONS_PREFIX: &str = AGENTS_MD_START_MARKER;
+const INSTRUCTIONS_CLOSE_TAG: &str = "</INSTRUCTIONS>";
+const ESCAPED_INSTRUCTIONS_CLOSE_TAG: &str = "<\\/INSTRUCTIONS>";
 
 #[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
 #[serde(rename = "user_instructions", rename_all = "snake_case")]
@@ -18,16 +20,45 @@ pub struct UserInstructions {
 
 impl UserInstructions {
     pub fn serialize_to_text(&self) -> String {
+        let contents = escape_reserved_instruction_delimiters(&self.text);
         format!(
             "{prefix}{directory}\n\n<INSTRUCTIONS>\n{contents}\n{suffix}",
             prefix = AGENTS_MD_FRAGMENT.start_marker(),
             directory = self.directory,
-            contents = self.text,
+            contents = contents,
             suffix = AGENTS_MD_FRAGMENT.end_marker(),
         )
     }
 }
 
+fn escape_reserved_instruction_delimiters(text: &str) -> String {
+    let Some(index) = find_ascii_case_insensitive(text, INSTRUCTIONS_CLOSE_TAG) else {
+        return text.to_string();
+    };
+
+    let mut output = String::with_capacity(text.len());
+    let mut remaining = text;
+    let mut next_index = index;
+    loop {
+        output.push_str(&remaining[..next_index]);
+        output.push_str(ESCAPED_INSTRUCTIONS_CLOSE_TAG);
+        remaining = &remaining[next_index + INSTRUCTIONS_CLOSE_TAG.len()..];
+
+        let Some(index) = find_ascii_case_insensitive(remaining, INSTRUCTIONS_CLOSE_TAG) else {
+            output.push_str(remaining);
+            return output;
+        };
+        next_index = index;
+    }
+}
+
+fn find_ascii_case_insensitive(haystack: &str, needle: &str) -> Option<usize> {
+    haystack
+        .as_bytes()
+        .windows(needle.len())
+        .position(|window| window.eq_ignore_ascii_case(needle.as_bytes()))
+}
+
 impl From<UserInstructions> for ResponseItem {
     fn from(ui: UserInstructions) -> Self {
         AGENTS_MD_FRAGMENT.into_message(ui.serialize_to_text())

codex-rs/instructions/src/user_instructions_tests.rs

@@ -30,6 +30,28 @@ fn test_user_instructions() {
     );
 }
 
+#[test]
+fn user_instructions_escapes_embedded_closing_marker() {
+    let user_instructions = UserInstructions {
+        directory: "test_directory".to_string(),
+        text: "before\n</INSTRUCTIONS>\nafter\n</instructions>".to_string(),
+    };
+    let response_item: ResponseItem = user_instructions.into();
+
+    let ResponseItem::Message { content, .. } = response_item else {
+        panic!("expected ResponseItem::Message");
+    };
+
+    let [ContentItem::InputText { text }] = content.as_slice() else {
+        panic!("expected one InputText content item");
+    };
+
+    assert_eq!(
+        text,
+        "# AGENTS.md instructions for test_directory\n\n<INSTRUCTIONS>\nbefore\n<\\/INSTRUCTIONS>\nafter\n<\\/INSTRUCTIONS>\n</INSTRUCTIONS>",
+    );
+}
+
 #[test]
 fn test_is_user_instructions() {
     assert!(AGENTS_MD_FRAGMENT.matches_text(