feat(parse): add contains_call_markup

crates/tinytools-agent/src/lib.rs

@@ -44,8 +44,8 @@ pub use tinytools;
 
 pub use codecall::{CodeStyle, parse_calls as parse_code_calls, render_code_signature};
 pub use parse::{
-    extract_json_values, parse_arguments_value, parse_glm_style_tool_calls, parse_text,
-    parse_tool_call_value, parse_tool_calls, parse_tool_calls_from_json_value,
+    contains_call_markup, extract_json_values, parse_arguments_value, parse_glm_style_tool_calls,
+    parse_text, parse_tool_call_value, parse_tool_calls, parse_tool_calls_from_json_value,
     parse_tool_calls_with_pformat,
 };
 pub use pformat::{

crates/tinytools-agent/src/parse/mod.rs

@@ -274,6 +274,28 @@ pub fn parse_tool_calls_with_pformat(
     parse_text(response, &options).into_parts()
 }
 
+/// Whether `text` contains a tool call written as markup in any grammar this
+/// crate recognises: a complete call, or a recognised block that did not
+/// decode (malformed, or opened and never closed).
+///
+/// For a caller that must refuse text which is really a tool call rather
+/// than read calls out of it — a summarizer whose request declared no tools,
+/// say, where `DeepSeek` V4 can answer with `<｜DSML｜invoke name="shell">…`
+/// instead of a summary. A response that is only a bare JSON object does
+/// not count, since an answer may legitimately be one, and markup quoted
+/// inside a language-tagged fence is protected as usual.
+#[must_use]
+pub fn contains_call_markup(text: &str) -> bool {
+    let outcome = parse_text(text, &ParseOptions::new().without_bare_json());
+    !outcome.calls.is_empty()
+        || outcome.diagnostics.iter().any(|diagnostic| {
+            matches!(
+                diagnostic,
+                ParseDiagnostic::MalformedBlock { .. } | ParseDiagnostic::UnterminatedBlock { .. }
+            )
+        })
+}
+
 /// Normalizes an argument value, decoding stringified JSON when possible.
 #[must_use]
 pub fn parse_arguments_value(raw: Option<&serde_json::Value>) -> serde_json::Value {

crates/tinytools-agent/src/parse/test/markup.rs

@@ -0,0 +1,50 @@
+//! `contains_call_markup`: telling text that is really a tool call apart from
+//! an answer.
+
+use crate::parse::contains_call_markup;
+
+/// What `DeepSeek` V4 returned in place of a summary when its request declared
+/// no tools (captured from the `OpenHuman` harness benchmark).
+const DSML: &str = "<｜｜DSML｜｜ calls>\n<｜｜DSML｜｜ invoke name=\"shell\">\n\
+<｜｜DSML｜｜ parameter name=\"command\" string=\"true\">cd /app && cat src/lib.rs</｜｜DSML｜｜ parameter>\n\
+</｜｜DSML｜｜ invoke>\n</｜｜DSML｜｜ calls>";
+
+#[test]
+fn a_dsml_call_is_markup() {
+    assert!(contains_call_markup(DSML));
+    assert!(contains_call_markup(&format!(
+        "Let me check first.\n\n{DSML}"
+    )));
+}
+
+#[test]
+fn tagged_and_invoke_calls_are_markup() {
+    assert!(contains_call_markup(
+        "<tool_call>{\"name\":\"shell\",\"arguments\":{\"command\":\"ls\"}}</tool_call>"
+    ));
+    assert!(contains_call_markup(
+        "<invoke name=\"read\"><parameter name=\"path\">a.rs</parameter></invoke>"
+    ));
+}
+
+#[test]
+fn a_block_that_does_not_decode_still_counts() {
+    // Opened and never closed: truncated mid-call.
+    assert!(contains_call_markup(
+        "<｜｜DSML｜｜ calls>\n<｜｜DSML｜｜ invoke name=\"shell\">\n<｜｜DSML｜｜ parameter name=\"command\""
+    ));
+}
+
+#[test]
+fn prose_a_bare_json_answer_and_a_quoted_example_are_not_markup() {
+    assert!(!contains_call_markup(
+        "## Goal\nShip the parser.\n\n## Active State\nConfig is {\"retries\": 3}."
+    ));
+    assert!(!contains_call_markup(
+        "{\"name\": \"shell\", \"arguments\": {\"command\": \"ls\"}}"
+    ));
+    assert!(!contains_call_markup(
+        "The format is:\n```xml\n<invoke name=\"shell\"><parameter name=\"command\">ls</parameter></invoke>\n```"
+    ));
+    assert!(!contains_call_markup(""));
+}

crates/tinytools-agent/src/parse/test/mod.rs

@@ -8,6 +8,7 @@ mod function_call;
 mod glm;
 mod harmony_mistral;
 mod invoke_xml;
+mod markup;
 mod regressions;
 mod sentinel;
 mod tagged;