patch 6.1

RETRIEX_PATCH_6_1_AUDIT_GAP_CLEANUP_README.md

@@ -0,0 +1,31 @@
+# RetrieX Patch 6.1 - Audit Gap Cleanup
+
+Scope: close the remaining config-source audit gaps that still reported `fallback_accessors_missing_yaml` after Patch 6.
+
+This patch intentionally does not remove PHP fallbacks yet and does not change runtime logic. It only adds the missing YAML keys with values copied 1:1 from the current PHP defaults so the following accessors have an explicit YAML source:
+
+- `AgentRunnerConfig::getOptimizedShopQueryTrimCharacters()`
+- `NdjsonHybridRetrieverConfig` retrieval vocabulary lists
+- `PromptBuilderConfig::getTechnicalProductKeywords()`
+- `PromptBuilderConfig::getAccessoryRequestKeywords()`
+
+Changed files:
+
+- `config/retriex/agent.yaml`
+- `config/retriex/retrieval.yaml`
+- `config/retriex/prompt.yaml`
+
+Recommended checks after applying:
+
+```bash
+php bin/console cache:clear
+php bin/console mto:agent:config:validate
+php bin/console mto:agent:config:audit-source --details
+php bin/console mto:agent:regression:test
+```
+
+Expected audit improvement:
+
+- `fallback_accessors_missing_yaml` should drop from `12` to `0`.
+
+`constructor_defaults_without_yaml_mapping` may still remain and should be handled separately before strict validation is enabled.

config/retriex/agent.yaml

@@ -6,6 +6,7 @@ parameters:
     product_search_knowledge_chunk_limit: 6
     advisory_product_search_knowledge_chunk_limit: 9
     optimized_shop_query_prefix_pattern: '/^(?:keywords?|suchquery|search\s*query|query)\s*:\s*/iu'
+    optimized_shop_query_trim_characters: " \t\n\r\0\x0B\"'`"
 
     messages:
       empty_prompt: '❌ Empty prompt.'

config/retriex/prompt.yaml

@@ -53,6 +53,71 @@ parameters:
         role_compatibility_label: Role compatibility with request
         role_incompatible_commercial_suppression_note: 'Commercial fields suppressed: this shop record is not a matching main-device result for the requested product role.'
     technical_product_keyword_match_threshold: 2
+    technical_product_keywords:
+      - technisch
+      - technical
+      - produkt
+      - product
+      - gerät
+      - device
+      - modell
+      - model
+      - messprinzip
+      - measurement principle
+      - schnittstelle
+      - interface
+      - relais
+      - relay
+      - indikator
+      - indicator
+      - grenzwert
+      - threshold
+      - messbereich
+      - measurement range
+      - gemessen
+      - measured
+      - minimaler
+      - minimum
+      - resthärte
+      - resthaerte
+      - °dh
+      - dh
+      - spannung
+      - voltage
+      - strom
+      - current
+      - druck
+      - pressure
+      - temperatur
+      - temperature
+      - schutzart
+      - ip
+      - fehlercode
+      - error code
+      - wasserhärte
+      - hardness
+      - testomat
+      - chlor
+      - chlormessung
+    accessory_request_keywords:
+      - passend
+      - passende
+      - passendes
+      - zubehör
+      - zubehor
+      - dazu
+      - indikator
+      - indikatoren
+      - ph-indikator
+      - ph indikator
+      - ph-indikatoren
+      - ph indikatoren
+      - reagenz
+      - kit
+      - set
+      - zusatz
+      - ergänzung
+      - ergaenzung
     sections:
       system_label: SYSTEM
       user_question_label: USER QUESTION

config/retriex/retrieval.yaml

@@ -26,6 +26,139 @@ parameters:
     focused_product_min_score: 10.0
     focused_product_min_gap: 4.0
     focused_product_max_chunks: 4
+    generic_product_tokens:
+      - produkt
+      - produkte
+      - produktkarte
+      - titel
+      - geraet
+      - gerät
+      - messgeraet
+      - messgerät
+      - wasser
+      - haerte
+      - härte
+      - resthaerte
+      - resthärte
+      - analyse
+      - analysator
+      - automat
+      - online
+      - messung
+      - messen
+      - preis
+      - preise
+      - kosten
+      - info
+      - infos
+      - passend
+      - richtige
+      - richtiges
+      - geeignet
+      - geeignete
+      - welche
+      - welcher
+      - welches
+      - brauche
+      - suche
+    important_short_model_tokens:
+      - th
+      - tc
+      - tp
+      - tm
+      - ph
+      - rx
+    family_descriptor_tokens:
+      - evo
+      - eco
+      - self
+      - clean
+      - mini
+      - pro
+      - plus
+      - basic
+      - lab
+      - inline
+      - compact
+      - panel
+      - sc
+    looks_like_reagent_tokens:
+      - indikator
+      - reagenz
+      - reagens
+      - laborchemikalie
+      - chemikalie
+      - sicherheitsdatenblatt
+      - sdb
+      - msds
+      - ufi
+      - gebinde
+      - flasche
+      - ersatzteil
+      - zubehoer
+      - zubehör
+      - service set
+      - filtereinsatz
+      - kerzenfilter
+      - druckregler
+    looks_like_safety_docs:
+      - sicherheitsdatenblatt
+      - sdb
+      - msds
+      - gefahrenbewertung
+      - gefahrenpiktogramm
+      - signalwort
+      - lagerung
+      - transport
+      - clp
+      - kennzeichnung
+      - h290
+      - pbt
+      - vpvb
+    looks_like_reagent_words:
+      - indikator
+      - reagenz
+      - reagens
+      - chemie
+      - chemikalie
+      - sdb
+      - sicherheitsdatenblatt
+      - msds
+      - flasche
+      - gebinde
+    looks_like_document_words:
+      - datenblatt
+      - dokument
+      - pdf
+      - handbuch
+      - manual
+      - beschreibung
+      - sdb
+      - sicherheitsdatenblatt
+      - msds
+    looks_like_safety_words:
+      - gefahr
+      - gefahrgut
+      - clp
+      - h290
+      - sicherheit
+      - kennzeichnung
+      - transport
+      - lagerung
+      - piktogramm
+    looks_like_device_words:
+      - geraet
+      - gerät
+      - messgeraet
+      - messgerät
+      - analysator
+      - automat
+      - messung
+      - messen
+      - ueberwachung
+      - überwachung
+      - online
+      - monitor
 
     # Vocabulary-backed retrieval token lists live in config/retriex/vocabulary.yaml.
     # The old per-key entries may still be added here to override a specific view.