feat: implement code fence awareness in tool sieve to prevent false-positive XML tool detection inside code blocks and refine prompt instructions.

This commit is contained in:
CJACK
2026-04-20 00:13:14 +08:00
parent 12256ceb24
commit f313d0068f
9 changed files with 300 additions and 143 deletions

View File

@@ -42,8 +42,8 @@ function consumeXMLToolCapture(captured, toolNames, trimWrappingJSONFence) {
suffix: trimmedFence.suffix,
};
}
// XML tool syntax but failed to parse — consume to avoid leak.
return { ready: true, prefix: prefixPart, calls: [], suffix: suffixPart };
// If this block failed to become a tool call, pass it through as text.
return { ready: true, prefix: prefixPart + xmlBlock, calls: [], suffix: suffixPart };
}
return { ready: false, prefix: '', calls: [], suffix: '' };
}
@@ -79,22 +79,8 @@ function findPartialXMLToolTagStart(s) {
return -1;
}
function looksLikeXMLToolTagFragment(s) {
const trimmed = (s || '').trim();
if (!trimmed) return false;
const lower = trimmed.toLowerCase();
const fragments = [
'tool_calls>', 'tool_call>', '/tool_calls>', '/tool_call>',
'function_calls>', 'function_call>', '/function_calls>', '/function_call>',
'invoke>', '/invoke>', 'tool_use>', '/tool_use>',
'tool_name>', '/tool_name>', 'parameters>', '/parameters>',
];
return fragments.some(f => lower.includes(f));
}
module.exports = {
consumeXMLToolCapture,
hasOpenXMLToolTag,
findPartialXMLToolTagStart,
looksLikeXMLToolTagFragment,
};

View File

@@ -12,7 +12,6 @@ const {
consumeXMLToolCapture: consumeXMLToolCaptureImpl,
hasOpenXMLToolTag,
findPartialXMLToolTagStart,
looksLikeXMLToolTagFragment,
} = require('./sieve-xml');
function processToolSieveChunk(state, chunk, toolNames) {
if (!state) {
@@ -77,7 +76,7 @@ function processToolSieveChunk(state, chunk, toolNames) {
resetIncrementalToolState(state);
continue;
}
const [safe, hold] = splitSafeContentForToolDetection(pending);
const [safe, hold] = splitSafeContentForToolDetection(state, pending);
if (!safe) {
break;
}
@@ -114,26 +113,22 @@ function flushToolSieve(state, toolNames) {
}
} else if (state.capture) {
const content = state.capture;
if (!hasOpenXMLToolTag(content) && !looksLikeXMLToolTagFragment(content)) {
noteText(state, content);
events.push({ type: 'text', text: content });
}
noteText(state, content);
events.push({ type: 'text', text: content });
}
state.capture = '';
state.capturing = false;
resetIncrementalToolState(state);
}
if (state.pending) {
if (!hasOpenXMLToolTag(state.pending) && !looksLikeXMLToolTagFragment(state.pending)) {
noteText(state, state.pending);
events.push({ type: 'text', text: state.pending });
}
noteText(state, state.pending);
events.push({ type: 'text', text: state.pending });
state.pending = '';
}
return events;
}
function splitSafeContentForToolDetection(s) {
function splitSafeContentForToolDetection(state, s) {
const text = s || '';
if (!text) {
return ['', ''];
@@ -141,6 +136,9 @@ function splitSafeContentForToolDetection(s) {
// Only hold back partial XML tool tags.
const xmlIdx = findPartialXMLToolTagStart(text);
if (xmlIdx >= 0) {
if (insideCodeFenceWithState(state, text.slice(0, xmlIdx))) {
return [text, ''];
}
if (xmlIdx > 0) {
return [text.slice(0, xmlIdx), text.slice(xmlIdx)];
}