marek/MtoRagSystem
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

harden history find tokens and shops earch

Browse Source
This commit is contained in:
team 1
2026-04-26 19:59:31 +02:00
parent ff273ff9a0
commit 308e980b4a
11 changed files with 1230 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

285
src/Config/AgentRunnerConfig.php
View File

@@ -46,6 +46,29 @@ final class AgentRunnerConfig
return is_numeric($value) ? (int) $value : $default;
}
private function getBool(string $key, bool $default): bool
{
$value = $this->value($key, $default);
if (is_bool($value)) {
return $value;
}
if (is_scalar($value)) {
$normalized = strtolower(trim((string) $value));
if (in_array($normalized, ['1', 'true', 'yes', 'on'], true)) {
return true;
}
if (in_array($normalized, ['0', 'false', 'no', 'off'], true)) {
return false;
}
}
return $default;
}
private function getString(string $key, string $default): string
{
$value = $this->value($key, $default);
@@ -122,6 +145,14 @@ final class AgentRunnerConfig
return $this->getString('messages.optimize_search', 'Ich optimiere die Recherche...');
}
public function getNoConcreteShopQueryMessage(): string
{
return $this->getString(
'messages.no_concrete_shop_query',
'Ich habe keine konkrete Shop-Suchanfrage erkannt. Bitte nenne das Produkt, Zubehör oder die Artikelnummer.'
);
}
public function getFetchSearchDataMessageTemplate(): string
{
return $this->getString('messages.fetch_search_data_template', 'Ich rufe Recherchedaten ab (type: %s)');
@@ -252,11 +283,14 @@ final class AgentRunnerConfig
'- Maximum 6 search terms, preferably fewer.',
'- Remove filler words, polite phrases, and irrelevant words.',
'- Preserve product names, brands, model numbers, and compound terms exactly if they are relevant.',
'- Preserve the language of the CURRENT USER INPUT for generic product/search terms; do not translate German search terms into English.',
'- For German user input, output German shop terms, for example "freies Chlor Messung" instead of "free chlorine measurement".',
'- Preserve domain terms from the current user input or resolved context in their original language.',
'- Numbers that belong to a product name or model must be preserved (e.g. Indikator 300, Testomat 808, Testomat 2000).',
'- Separate terms using spaces only.',
'- If a relevant product name is present, it must be placed at the beginning of the final search query.',
'- Try to always identify all products mentioned in the user input text, even in long prompts.',
'- Look for terms such as Testomat, Horiba, Tritromat, or words like indicator.',
'- Look for terms such as Testomat, Horiba, Tritromat, or words like indicator/Indikator.',
'- If the current user input is vague or referential, use the recent conversation context only as support.',
'- Do not output words that only describe conversation flow, such as "same", "again", "also", or "like above".',
]);
@@ -297,6 +331,253 @@ final class AgentRunnerConfig
return $this->getString('shop_prompt.current_user_input_label', 'CURRENT USER INPUT');
}
public function isShopQueryLanguagePreservationEnabled(): bool
{
return $this->getBool('shop_prompt.language_preservation.enabled', true);
}
/**
* @return array<string, string[]>
*/
public function getShopQueryLanguageMarkers(): array
{
$default = [
'de' => [
' ä ', ' ö ', ' ü ', ' ß ',
' der ', ' die ', ' das ', ' ein ', ' eine ', ' einer ', ' einen ',
' welchem ', ' welchen ', ' welche ', ' welcher ',
' kann ', ' nutzen ', ' zur ', ' für ', ' fuer ',
' messung ', ' indikator ', ' reagenz ', ' chlor ',
],
];
$value = $this->value('shop_prompt.language_preservation.language_markers', $default);
if (!is_array($value)) {
return $default;
}
$out = [];
foreach ($value as $language => $markers) {
if (!is_string($language) || !is_array($markers)) {
continue;
}
$cleanMarkers = [];
foreach ($markers as $marker) {
if (!is_scalar($marker)) {
continue;
}
$marker = strtolower((string) $marker);
if ($marker !== '') {
$cleanMarkers[] = $marker;
}
}
if ($cleanMarkers !== []) {
$out[$language] = array_values(array_unique($cleanMarkers));
}
}
return $out !== [] ? $out : $default;
}
/**
* @return array<string, string>
*/
public function isShopQueryMetaGuardEnabled(): bool
{
return $this->getBool('shop_prompt.meta_query_guard.enabled', true);
}
/**
* @return string[]
*/
public function getShopQueryMetaOnlyTerms(): array
{
return $this->getStringList('shop_prompt.meta_query_guard.meta_only_terms', [
'shop',
'shopsuche',
'shop-suche',
'suche',
'suchen',
'such',
'finde',
'find',
'zeige',
'zeig',
'bitte',
'mal',
'im',
'in',
'nach',
'den',
'die',
'das',
'der',
'dem',
]);
}
public function isShopQueryContextFallbackEnabled(): bool
{
return $this->getBool('shop_prompt.meta_query_guard.context_fallback_enabled', true);
}
public function getShopQueryContextFallbackQuestionLimit(): int
{
return $this->getInt('shop_prompt.meta_query_guard.context_fallback_question_limit', 12);
}
public function getShopQueryContextFallbackHistoryBudgetChars(): int
{
return $this->getInt('shop_prompt.meta_query_guard.context_fallback_history_budget_chars', 20000);
}
public function shouldUseFullHistoryForShopQueryContextFallback(): bool
{
return $this->getBool('shop_prompt.meta_query_guard.context_fallback_use_full_history', true);
}
public function getShopQueryContextFallbackMaxTerms(): int
{
return $this->getInt('shop_prompt.meta_query_guard.context_fallback_max_terms', 6);
}
/**
* @return string[]
*/
public function getShopQueryContextFallbackFilterTerms(): array
{
return $this->getStringList('shop_prompt.meta_query_guard.context_fallback_filter_terms', [
'mit',
'welche',
'welcher',
'welches',
'welchem',
'welchen',
'was',
'wie',
'wo',
'kann',
'koennen',
'können',
'konnte',
'könnte',
'ich',
'wir',
'man',
'nutzen',
'benutzen',
'verwenden',
'verwende',
'nehmen',
'zur',
'zum',
'für',
'fuer',
'messen',
'gemessen',
'messung',
]);
}
public function isShopQueryContextAnchorEnrichmentEnabled(): bool
{
return $this->getBool('shop_prompt.context_anchor_enrichment.enabled', true);
}
public function getShopQueryContextAnchorEnrichmentMaxQueryTerms(): int
{
return $this->getInt('shop_prompt.context_anchor_enrichment.max_query_terms', 2);
}
/**
* @return string[]
*/
public function getShopQueryContextAnchorEnrichmentTriggerTerms(): array
{
return $this->getStringList('shop_prompt.context_anchor_enrichment.trigger_terms', [
'indikator',
'indikatortyp',
'indicator',
'reagenz',
'reagenzsatz',
'reagent',
'zubehör',
'zubehor',
'accessory',
]);
}
/**
* @return string[]
*/
public function getShopQueryContextAnchorEnrichmentPatterns(): array
{
return $this->getStringList('shop_prompt.context_anchor_enrichment.anchor_patterns', [
'/\b(?:indikator(?:typ)?|indicator(?:\s+type)?|reagenz(?:satz|typ)?|reagent(?:\s+set|\s+type)?|typ|type)\s+[A-Za-zÄÖÜäöüß]{0,8}\s*\d{1,5}(?:\s*[A-ZÄÖÜ]{1,4})?(?:\s*%)?\b/iu',
]);
}
public function getShopQueryContextAnchorEnrichmentTemplate(): string
{
return $this->getString('shop_prompt.context_anchor_enrichment.template', '{anchor} {query}');
}
public function getShopQueryTranslationReplacements(string $language): array
{
$default = [
'de' => [
'free chlorine' => 'freies chlor',
'free chlor' => 'freies chlor',
'total chlorine' => 'gesamtchlor',
'chlorine measurement' => 'chlor messung',
'water hardness' => 'wasserhärte',
'measurement' => 'messung',
'measuring' => 'messung',
'chlorine' => 'chlor',
'indicator' => 'indikator',
'indicators' => 'indikatoren',
'reagent' => 'reagenz',
'reagents' => 'reagenzien',
'accessory' => 'zubehör',
'accessories' => 'zubehör',
],
];
$value = $this->value(
'shop_prompt.language_preservation.translation_replacements.' . $language,
$default[$language] ?? []
);
if (!is_array($value)) {
return $default[$language] ?? [];
}
$out = [];
foreach ($value as $source => $target) {
if (!is_scalar($source) || !is_scalar($target)) {
continue;
}
$source = strtolower(trim((string) $source));
$target = trim((string) $target);
if ($source !== '' && $target !== '') {
$out[$source] = $target;
}
}
uksort($out, static fn(string $a, string $b): int => strlen($b) <=> strlen($a));
return $out !== [] ? $out : ($default[$language] ?? []);
}
private function buildRulesBlock(array $rules, string $headline = 'Rules:'): string
{
return $headline . "\n" . implode("\n", $rules);
@@ -317,4 +598,4 @@ final class AgentRunnerConfig
return implode("\n\n", $normalized);
}
}
}

73
src/Config/RetriexEffectiveConfigProvider.php
View File

@@ -192,6 +192,47 @@ final readonly class RetriexEffectiveConfigProvider
$errors[] = 'Shop query optimizer prompt no longer contains the original query.';
}
$metaOnlyTerms = $this->agentRunnerConfig->getShopQueryMetaOnlyTerms();
foreach (['shop', 'suche'] as $term) {
$key = 'shop_query_meta_guard_term_' . $term;
$checks[$key] = in_array($term, $metaOnlyTerms, true);
if (!$checks[$key]) {
$errors[] = 'Missing shop query meta guard term: ' . $term;
}
}
$checks['shop_query_context_fallback_enabled'] = $this->agentRunnerConfig->isShopQueryContextFallbackEnabled();
if (!$checks['shop_query_context_fallback_enabled']) {
$errors[] = 'Shop query context fallback is disabled.';
}
$contextFallbackFilterTerms = $this->agentRunnerConfig->getShopQueryContextFallbackFilterTerms();
foreach (['welchem', 'kann', 'messen'] as $term) {
$key = 'shop_query_context_fallback_filter_' . $term;
$checks[$key] = in_array($term, $contextFallbackFilterTerms, true);
if (!$checks[$key]) {
$errors[] = 'Missing shop query context fallback filter term: ' . $term;
}
}
$checks['shop_query_context_fallback_history_budget_positive'] = $this->agentRunnerConfig->getShopQueryContextFallbackHistoryBudgetChars() > 0;
if (!$checks['shop_query_context_fallback_history_budget_positive']) {
$errors[] = 'Shop query context fallback history budget must be greater than zero.';
}
$checks['shop_query_context_fallback_full_history_enabled'] = $this->agentRunnerConfig->shouldUseFullHistoryForShopQueryContextFallback();
if (!$checks['shop_query_context_fallback_full_history_enabled']) {
$errors[] = 'Shop query context fallback full-history fallback is disabled.';
}
$checks['shop_query_context_fallback_question_limit_minimum'] = $this->agentRunnerConfig->getShopQueryContextFallbackQuestionLimit() >= 6;
if (!$checks['shop_query_context_fallback_question_limit_minimum']) {
$errors[] = 'Shop query context fallback question limit is too low for repeated meta follow-ups.';
}
$checks['shop_query_context_fallback_max_terms_positive'] = $this->agentRunnerConfig->getShopQueryContextFallbackMaxTerms() > 0;
if (!$checks['shop_query_context_fallback_max_terms_positive']) {
$errors[] = 'Shop query context fallback max terms must be greater than zero.';
}
$status = $errors === [] ? 'OK' : 'ERROR';
return [
@@ -362,6 +403,7 @@ final readonly class RetriexEffectiveConfigProvider
'check_internet_sources' => $this->agentRunnerConfig->getCheckInternetSourcesMessage(),
'retrieve_knowledge' => $this->agentRunnerConfig->getRetrieveKnowledgeMessage(),
'optimize_search' => $this->agentRunnerConfig->getOptimizeSearchMessage(),
'no_concrete_shop_query' => $this->agentRunnerConfig->getNoConcreteShopQueryMessage(),
'fetch_search_data_template' => $this->agentRunnerConfig->getFetchSearchDataMessageTemplate(),
'analyze_all_information' => $this->agentRunnerConfig->getAnalyzeAllInformationMessage(),
'thinking_while_streaming' => $this->agentRunnerConfig->getThinkingWhileStreamingMessage(),
@@ -392,6 +434,28 @@ final readonly class RetriexEffectiveConfigProvider
'output_format_block' => $this->agentRunnerConfig->getShopPromptOutputFormatBlock(),
'recent_conversation_context_label' => $this->agentRunnerConfig->getRecentConversationContextLabel(),
'current_user_input_label' => $this->agentRunnerConfig->getCurrentUserInputLabel(),
'language_preservation' => [
'enabled' => $this->agentRunnerConfig->isShopQueryLanguagePreservationEnabled(),
'language_markers' => $this->agentRunnerConfig->getShopQueryLanguageMarkers(),
'translation_replacements_de' => $this->agentRunnerConfig->getShopQueryTranslationReplacements('de'),
],
'context_anchor_enrichment' => [
'enabled' => $this->agentRunnerConfig->isShopQueryContextAnchorEnrichmentEnabled(),
'max_query_terms' => $this->agentRunnerConfig->getShopQueryContextAnchorEnrichmentMaxQueryTerms(),
'trigger_terms' => $this->agentRunnerConfig->getShopQueryContextAnchorEnrichmentTriggerTerms(),
'anchor_patterns' => $this->agentRunnerConfig->getShopQueryContextAnchorEnrichmentPatterns(),
'template' => $this->agentRunnerConfig->getShopQueryContextAnchorEnrichmentTemplate(),
],
'meta_query_guard' => [
'enabled' => $this->agentRunnerConfig->isShopQueryMetaGuardEnabled(),
'context_fallback_use_full_history' => $this->agentRunnerConfig->shouldUseFullHistoryForShopQueryContextFallback(),
'meta_only_terms' => $this->agentRunnerConfig->getShopQueryMetaOnlyTerms(),
'context_fallback_enabled' => $this->agentRunnerConfig->isShopQueryContextFallbackEnabled(),
'context_fallback_question_limit' => $this->agentRunnerConfig->getShopQueryContextFallbackQuestionLimit(),
'context_fallback_history_budget_chars' => $this->agentRunnerConfig->getShopQueryContextFallbackHistoryBudgetChars(),
'context_fallback_max_terms' => $this->agentRunnerConfig->getShopQueryContextFallbackMaxTerms(),
'context_fallback_filter_terms' => $this->agentRunnerConfig->getShopQueryContextFallbackFilterTerms(),
],
],
];
}
@@ -795,6 +859,15 @@ final readonly class RetriexEffectiveConfigProvider
$this->validateStringListMap($agent['html_templates'] ?? [], 'agent.html_templates', $errors, $warnings);
$this->validateStringListMap($agent['shop_query_optimizer'] ?? [], 'agent.shop_query_optimizer', $errors, $warnings);
$this->validateRegexPattern($agent['optimized_shop_query_prefix_pattern'] ?? null, 'agent.optimized_shop_query_prefix_pattern', $errors);
$anchorEnrichment = $agent['shop_query_optimizer']['context_anchor_enrichment'] ?? [];
if (is_array($anchorEnrichment)) {
$this->validateStringList($this->toList($anchorEnrichment['trigger_terms'] ?? []), 'agent.shop_query_optimizer.context_anchor_enrichment.trigger_terms', $errors, $warnings);
$this->validateRegexPatternList($anchorEnrichment['anchor_patterns'] ?? [], 'agent.shop_query_optimizer.context_anchor_enrichment.anchor_patterns', $errors);
if (trim((string) ($anchorEnrichment['template'] ?? '')) === '') {
$errors[] = 'agent.shop_query_optimizer.context_anchor_enrichment.template must not be empty.';
}
}
}
/**