diff --git a/docs/prompt-compatibility.md b/docs/prompt-compatibility.md
index 6039e34..c7a6ad4 100644
--- a/docs/prompt-compatibility.md
+++ b/docs/prompt-compatibility.md
@@ -153,6 +153,7 @@ OpenAI Chat / Responses 在标准化后、history split / current input file 之
工具调用正例现在优先示范官方 DSML 风格：`<|DSML|tool_calls>` → `<|DSML|invoke name="...">` → `<|DSML|parameter name="...">`。
兼容层仍接受旧式纯 `` wrapper，但提示词会优先要求模型输出官方 DSML 标签，并强调不能只输出 closing wrapper 而漏掉 opening tag。需要注意：这是“兼容 DSML 外壳，内部仍以 XML 解析语义为准”，不是原生 DSML 全链路实现；DSML 标签会在解析入口归一化回现有 XML 标签后继续走同一套 parser。
+数组参数使用 `- ...
` 子节点表示；当某个参数体只包含 item 子节点时，Go / Node 解析器会把它还原成数组，避免 `questions` / `options` 这类 schema 中要求 array 的参数被误解析成 `{ "item": ... }` 对象。若模型把完整结构化 XML fragment 误包进 CDATA，兼容层会在保护 `content` / `command` 等原文字段的前提下，尝试把非原文字段中的 CDATA XML fragment 还原成 object / array。
正例中的工具名只会来自当前请求实际声明的工具；如果当前请求没有足够的已知工具形态，就省略对应的单工具、多工具或嵌套示例，避免把不可用工具名写进 prompt。
对执行类工具，脚本内容必须进入执行参数本身：`Bash` / `execute_command` 使用 `command`，`exec_command` 使用 `cmd`；不要把脚本示范成 `path` / `content` 文件写入参数。
diff --git a/docs/toolcall-semantics.md b/docs/toolcall-semantics.md
index 5529a4b..bb7b924 100644
--- a/docs/toolcall-semantics.md
+++ b/docs/toolcall-semantics.md
@@ -63,6 +63,8 @@
- 当文本中 mention 了某种标签名（如 `` 或 Markdown inline code 里的 `<|DSML|tool_calls>`）而后面紧跟真正工具调用时，sieve 会跳过不可解析的 mention 候选并继续匹配后续真实工具块，不会因 mention 导致工具调用丢失，也不会截断 mention 后的正文
另外，`` 的值如果本身是合法 JSON 字面量，也会按结构化值解析，而不是一律保留为字符串。例如 `123`、`true`、`null`、`[1,2]`、`{"a":1}` 都会还原成对应的 number / boolean / null / array / object。
+结构化 XML 参数也会还原为 JSON 结构：如果参数体只包含一个或多个 `- ...
` 子节点，会输出数组；嵌套对象里的 item-only 字段也同样按数组处理。例如 `- ...
` 会输出 `{"questions":[{"question":"..."}]}`，而不是 `{"questions":{"item":...}}`。
+如果模型误把完整结构化 XML fragment 放进 CDATA，Go / Node 会先保护明显的原文字段（如 `content` / `command` / `prompt` / `old_string` / `new_string`），其余参数会尝试把 CDATA 内的完整 XML fragment 还原成 object / array；常见的 `
` 分隔符会按换行归一化后再解析。
## 4) 输出结构
diff --git a/internal/js/helpers/stream-tool-sieve/parse_payload.js b/internal/js/helpers/stream-tool-sieve/parse_payload.js
index 185ed4d..a935f00 100644
--- a/internal/js/helpers/stream-tool-sieve/parse_payload.js
+++ b/internal/js/helpers/stream-tool-sieve/parse_payload.js
@@ -1,6 +1,5 @@
'use strict';
-const TOOL_CALL_MARKUP_KV_PATTERN = /<(?:[a-z0-9_:-]+:)?([a-z0-9_.-]+)\b[^>]*>([\s\S]*?)<\/(?:[a-z0-9_:-]+:)?\1>/gi;
const CDATA_PATTERN = /^$/i;
const XML_ATTR_PATTERN = /\b([a-z0-9_:-]+)\s*=\s*("([^"]*)"|'([^']*)')/gi;
const TOOL_MARKUP_NAMES = ['tool_calls', 'invoke', 'parameter'];
@@ -293,7 +292,7 @@ function parseMarkupSingleToolCall(block) {
if (!paramName) {
continue;
}
- appendMarkupValue(input, paramName, parseMarkupValue(match.body));
+ appendMarkupValue(input, paramName, parseMarkupValue(match.body, paramName));
}
if (Object.keys(input).length === 0 && inner.trim() !== '') {
return null;
@@ -600,8 +599,11 @@ function parseMarkupInput(raw) {
return {};
}
// Prioritize XML-style KV tags (e.g., val)
- const kv = parseMarkupKVObject(s);
- if (Object.keys(kv).length > 0) {
+ const kv = unwrapItemOnlyMarkupValue(parseMarkupKVObject(s));
+ if (Array.isArray(kv)) {
+ return kv;
+ }
+ if (kv && typeof kv === 'object' && Object.keys(kv).length > 0) {
return kv;
}
@@ -622,12 +624,12 @@ function parseMarkupKVObject(text) {
return {};
}
const out = {};
- for (const m of raw.matchAll(TOOL_CALL_MARKUP_KV_PATTERN)) {
- const key = toStringSafe(m[1]).trim();
+ for (const block of findGenericXmlElementBlocks(raw)) {
+ const key = toStringSafe(block.localName).trim();
if (!key) {
continue;
}
- const value = parseMarkupValue(m[2]);
+ const value = parseMarkupValue(block.body, key);
if (value === undefined || value === null) {
continue;
}
@@ -636,11 +638,146 @@ function parseMarkupKVObject(text) {
return out;
}
-function parseMarkupValue(raw) {
+function findGenericXmlElementBlocks(text) {
+ const source = toStringSafe(text);
+ if (!source) {
+ return [];
+ }
+ const out = [];
+ let pos = 0;
+ while (pos < source.length) {
+ const start = findGenericXmlStartTagOutsideCDATA(source, pos);
+ if (!start) {
+ break;
+ }
+ if (start.selfClosing) {
+ out.push({
+ name: start.name,
+ localName: start.localName,
+ attrs: start.attrs,
+ body: '',
+ start: start.start,
+ end: start.end + 1,
+ });
+ pos = start.end + 1;
+ continue;
+ }
+ const end = findMatchingGenericXmlEndTagOutsideCDATA(source, start.name, start.bodyStart);
+ if (!end) {
+ pos = start.bodyStart;
+ continue;
+ }
+ out.push({
+ name: start.name,
+ localName: start.localName,
+ attrs: start.attrs,
+ body: source.slice(start.bodyStart, end.closeStart),
+ start: start.start,
+ end: end.closeEnd,
+ });
+ pos = end.closeEnd;
+ }
+ return out;
+}
+
+function findGenericXmlStartTagOutsideCDATA(text, from) {
+ const lower = text.toLowerCase();
+ for (let i = Math.max(0, from || 0); i < text.length;) {
+ const skipped = skipXmlIgnoredSection(lower, i);
+ if (skipped.blocked) {
+ return null;
+ }
+ if (skipped.advanced) {
+ i = skipped.next;
+ continue;
+ }
+ if (text[i] !== '<' || text[i + 1] === '/' || text[i + 1] === '!' || text[i + 1] === '?') {
+ i += 1;
+ continue;
+ }
+ const match = text.slice(i + 1).match(/^([A-Za-z_][A-Za-z0-9_.:-]*)/);
+ if (!match) {
+ i += 1;
+ continue;
+ }
+ const name = match[1];
+ const nameEnd = i + 1 + name.length;
+ if (!hasXmlTagBoundary(text, nameEnd)) {
+ i += 1;
+ continue;
+ }
+ const tagEnd = findXmlTagEnd(text, nameEnd);
+ if (tagEnd < 0) {
+ return null;
+ }
+ return {
+ start: i,
+ end: tagEnd,
+ bodyStart: tagEnd + 1,
+ name,
+ localName: name.includes(':') ? name.slice(name.lastIndexOf(':') + 1) : name,
+ attrs: text.slice(nameEnd, tagEnd),
+ selfClosing: isSelfClosingXmlTag(text.slice(i, tagEnd)),
+ };
+ }
+ return null;
+}
+
+function findMatchingGenericXmlEndTagOutsideCDATA(text, name, from) {
+ const lower = text.toLowerCase();
+ const needle = toStringSafe(name).toLowerCase();
+ if (!needle) {
+ return null;
+ }
+ const openTarget = `<${needle}`;
+ const closeTarget = `')) {
- const nested = parseMarkupInput(s);
- if (nested && typeof nested === 'object' && !Array.isArray(nested)) {
+ const nested = unwrapItemOnlyMarkupValue(parseMarkupInput(s));
+ if (Array.isArray(nested)) {
+ return nested;
+ }
+ if (nested && typeof nested === 'object') {
if (isOnlyRawValue(nested)) {
return toStringSafe(nested._raw);
}
@@ -664,6 +804,66 @@ function parseMarkupValue(raw) {
return s;
}
+function parseStructuredCDATAParameterValue(paramName, raw) {
+ if (preservesCDATAStringParameter(paramName)) {
+ return { ok: false, value: null };
+ }
+ const normalized = normalizeCDATAForStructuredParse(raw);
+ if (!normalized.includes('<') || !normalized.includes('>')) {
+ return { ok: false, value: null };
+ }
+ const parsed = parseMarkupInput(normalized);
+ if (Array.isArray(parsed)) {
+ return { ok: true, value: parsed };
+ }
+ if (parsed && typeof parsed === 'object' && !isOnlyRawValue(parsed) && Object.keys(parsed).length > 0) {
+ return { ok: true, value: parsed };
+ }
+ return { ok: false, value: null };
+}
+
+function normalizeCDATAForStructuredParse(raw) {
+ return unescapeHtml(toStringSafe(raw).replace(//gi, '\n').trim());
+}
+
+function preservesCDATAStringParameter(name) {
+ return new Set([
+ 'content',
+ 'file_content',
+ 'text',
+ 'prompt',
+ 'query',
+ 'command',
+ 'cmd',
+ 'script',
+ 'code',
+ 'old_string',
+ 'new_string',
+ 'pattern',
+ 'path',
+ 'file_path',
+ ]).has(toStringSafe(name).trim().toLowerCase());
+}
+
+function unwrapItemOnlyMarkupValue(value) {
+ if (Array.isArray(value)) {
+ return value.map(unwrapItemOnlyMarkupValue);
+ }
+ if (!value || typeof value !== 'object') {
+ return value;
+ }
+ const keys = Object.keys(value);
+ if (keys.length === 1 && keys[0] === 'item') {
+ const items = unwrapItemOnlyMarkupValue(value.item);
+ return Array.isArray(items) ? items : [items];
+ }
+ const out = {};
+ for (const key of keys) {
+ out[key] = unwrapItemOnlyMarkupValue(value[key]);
+ }
+ return out;
+}
+
function extractRawTagValue(inner) {
const s = toStringSafe(inner).trim();
if (!s) {
diff --git a/internal/toolcall/toolcalls_parse_markup.go b/internal/toolcall/toolcalls_parse_markup.go
index 8633ad0..d16f5e1 100644
--- a/internal/toolcall/toolcalls_parse_markup.go
+++ b/internal/toolcall/toolcalls_parse_markup.go
@@ -10,6 +10,7 @@ import (
var xmlAttrPattern = regexp.MustCompile(`(?is)\b([a-z0-9_:-]+)\s*=\s*("([^"]*)"|'([^']*)')`)
var xmlToolCallsClosePattern = regexp.MustCompile(`(?is)`)
var xmlInvokeStartPattern = regexp.MustCompile(`(?is)]*\bname\s*=\s*("([^"]*)"|'([^']*)')`)
+var cdataBRSeparatorPattern = regexp.MustCompile(`(?i)`)
func parseXMLToolCalls(text string) []ParsedToolCall {
wrappers := findXMLElementBlocks(text, "tool_calls")
@@ -91,7 +92,7 @@ func parseSingleXMLToolCall(block xmlElementBlock) (ParsedToolCall, bool) {
if paramName == "" {
continue
}
- value := parseInvokeParameterValue(paramMatch.Body)
+ value := parseInvokeParameterValue(paramName, paramMatch.Body)
appendMarkupValue(input, paramName, value)
}
@@ -289,7 +290,7 @@ func parseXMLTagAttributes(raw string) map[string]string {
return out
}
-func parseInvokeParameterValue(raw string) any {
+func parseInvokeParameterValue(paramName, raw string) any {
trimmed := strings.TrimSpace(raw)
if trimmed == "" {
return ""
@@ -298,10 +299,34 @@ func parseInvokeParameterValue(raw string) any {
if parsed, ok := parseJSONLiteralValue(value); ok {
return parsed
}
+ if parsed, ok := parseStructuredCDATAParameterValue(paramName, value); ok {
+ return parsed
+ }
return value
}
decoded := html.UnescapeString(extractRawTagValue(trimmed))
if strings.Contains(decoded, "<") && strings.Contains(decoded, ">") {
+ if parsedValue, ok := parseXMLFragmentValue(decoded); ok {
+ switch v := parsedValue.(type) {
+ case map[string]any:
+ if len(v) > 0 {
+ return v
+ }
+ case []any:
+ return v
+ case string:
+ text := strings.TrimSpace(v)
+ if text == "" {
+ return ""
+ }
+ if parsedText, ok := parseJSONLiteralValue(text); ok {
+ return parsedText
+ }
+ return v
+ default:
+ return v
+ }
+ }
if parsed := parseStructuredToolCallInput(decoded); len(parsed) > 0 {
if len(parsed) == 1 {
if rawValue, ok := parsed["_raw"].(string); ok {
@@ -316,3 +341,45 @@ func parseInvokeParameterValue(raw string) any {
}
return decoded
}
+
+func parseStructuredCDATAParameterValue(paramName, raw string) (any, bool) {
+ if preservesCDATAStringParameter(paramName) {
+ return nil, false
+ }
+ normalized := normalizeCDATAForStructuredParse(raw)
+ if !strings.Contains(normalized, "<") || !strings.Contains(normalized, ">") {
+ return nil, false
+ }
+ parsed, ok := parseXMLFragmentValue(normalized)
+ if !ok {
+ return nil, false
+ }
+ switch v := parsed.(type) {
+ case []any:
+ return v, true
+ case map[string]any:
+ if len(v) == 0 {
+ return nil, false
+ }
+ return v, true
+ default:
+ return nil, false
+ }
+}
+
+func normalizeCDATAForStructuredParse(raw string) string {
+ if raw == "" {
+ return ""
+ }
+ normalized := cdataBRSeparatorPattern.ReplaceAllString(raw, "\n")
+ return html.UnescapeString(strings.TrimSpace(normalized))
+}
+
+func preservesCDATAStringParameter(name string) bool {
+ switch strings.ToLower(strings.TrimSpace(name)) {
+ case "content", "file_content", "text", "prompt", "query", "command", "cmd", "script", "code", "old_string", "new_string", "pattern", "path", "file_path":
+ return true
+ default:
+ return false
+ }
+}
diff --git a/internal/toolcall/toolcalls_test.go b/internal/toolcall/toolcalls_test.go
index b48f88c..30a7b3a 100644
--- a/internal/toolcall/toolcalls_test.go
+++ b/internal/toolcall/toolcalls_test.go
@@ -159,6 +159,82 @@ func TestParseToolCallsSupportsJSONScalarParameters(t *testing.T) {
}
}
+func TestParseToolCallsTreatsItemOnlyParameterBodyAsArray(t *testing.T) {
+ text := strings.Join([]string{
+ `<|DSML|tool_calls>`,
+ `<|DSML|invoke name="AskUserQuestion">`,
+ `<|DSML|parameter name="questions">`,
+ `- `,
+ ``,
+ ``,
+ ``,
+ `
`,
+ ``,
+ ``,
+ ``,
+ }, "\n")
+ calls := ParseToolCalls(text, []string{"AskUserQuestion"})
+ if len(calls) != 1 {
+ t.Fatalf("expected one AskUserQuestion call, got %#v", calls)
+ }
+ questions, ok := calls[0].Input["questions"].([]any)
+ if !ok || len(questions) != 1 {
+ t.Fatalf("expected questions to parse as array, got %#v", calls[0].Input["questions"])
+ }
+ first, ok := questions[0].(map[string]any)
+ if !ok {
+ t.Fatalf("expected first question object, got %#v", questions[0])
+ }
+ if first["question"] != "What would you like to do next?" || first["header"] != "Next step" || first["multiSelect"] != false {
+ t.Fatalf("unexpected question payload: %#v", first)
+ }
+ options, ok := first["options"].([]any)
+ if !ok || len(options) != 2 {
+ t.Fatalf("expected options to parse as array, got %#v", first["options"])
+ }
+}
+
+func TestParseToolCallsTreatsCDATAItemOnlyBodyAsArray(t *testing.T) {
+ todos := `
Testing EnterWorktree tool
Test EnterWorktree tool
in_progress
Testing TodoWrite tool
Test TodoWrite tool
completed
`
+ text := `<|DSML|tool_calls><|DSML|invoke name="TodoWrite"><|DSML|parameter name="todos">`
+ calls := ParseToolCalls(text, []string{"TodoWrite"})
+ if len(calls) != 1 {
+ t.Fatalf("expected one TodoWrite call, got %#v", calls)
+ }
+ items, ok := calls[0].Input["todos"].([]any)
+ if !ok || len(items) != 2 {
+ t.Fatalf("expected todos CDATA item body to parse as array, got %#v", calls[0].Input["todos"])
+ }
+ first, ok := items[0].(map[string]any)
+ if !ok {
+ t.Fatalf("expected first todo object, got %#v", items[0])
+ }
+ if first["activeForm"] != "Testing EnterWorktree tool" || first["content"] != "Test EnterWorktree tool" || first["status"] != "in_progress" {
+ t.Fatalf("unexpected first todo: %#v", first)
+ }
+}
+
+func TestParseToolCallsTreatsCDATAObjectFragmentAsObject(t *testing.T) {
+ payload := `- ',
+ '',
+ '',
+ '',
+ '
',
+ '',
+ '',
+ '',
+ ].join('\n');
+ const calls = parseToolCalls(payload, ['AskUserQuestion']);
+ assert.equal(calls.length, 1);
+ assert.deepEqual(calls[0].input.questions, [
+ {
+ question: 'What would you like to do next?',
+ header: 'Next step',
+ options: [
+ { label: 'Run tests', description: 'Run the test suite' },
+ { label: 'Other task', description: 'Something else entirely' },
+ ],
+ multiSelect: false,
+ },
+ ]);
+});
+
+test('parseToolCalls treats CDATA item-only body as array', () => {
+ const todos = '
Testing EnterWorktree tool
Test EnterWorktree tool
in_progress
Testing TodoWrite tool
Test TodoWrite tool
completed
';
+ const payload = `<|DSML|tool_calls><|DSML|invoke name="TodoWrite"><|DSML|parameter name="todos">`;
+ const calls = parseToolCalls(payload, ['TodoWrite']);
+ assert.equal(calls.length, 1);
+ assert.deepEqual(calls[0].input.todos, [
+ {
+ activeForm: 'Testing EnterWorktree tool',
+ content: 'Test EnterWorktree tool',
+ status: 'in_progress',
+ },
+ {
+ activeForm: 'Testing TodoWrite tool',
+ content: 'Test TodoWrite tool',
+ status: 'completed',
+ },
+ ]);
+});
+
+test('parseToolCalls treats CDATA object fragment as object', () => {
+ const fragment = '