Track the practice cluster's Helm values in Git

.github/workflows/helm-chart.yml

@@ -0,0 +1,80 @@
+name: Validate Helm chart
+
+on:
+  push:
+    paths:
+      - infra/helm/**
+      - .github/workflows/helm-chart.yml
+  pull_request:
+    paths:
+      - infra/helm/**
+      - .github/workflows/helm-chart.yml
+
+permissions:
+  contents: read
+
+jobs:
+  validate:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: azure/setup-helm@v4
+        with:
+          version: v3.17.3
+      - name: Lint chart
+        run: helm lint infra/helm/medical-chatbot
+      - name: Render default and production manifests
+        run: |
+          helm template default infra/helm/medical-chatbot > /tmp/default.yaml
+          helm template production infra/helm/medical-chatbot \
+            --values infra/helm/medical-chatbot/values-prod.yaml \
+            > /tmp/production.yaml
+          grep -q 'ANSWER_MODEL_ID: "qwen.qwen3-next-80b-a3b"' /tmp/production.yaml
+          grep -q 'RERANK_ENABLED: "true"' /tmp/production.yaml
+          grep -q 'checksum/runtime-config:' /tmp/production.yaml
+
+      # The practice cluster is only evidence for the production migration
+      # while it renders the same behavioural contract as production, so both
+      # of its releases are asserted here rather than trusted by review.
+      - name: Render practice manifests
+        run: |
+          helm template medical-chatbot-app infra/helm/medical-chatbot \
+            --values infra/helm/medical-chatbot/values-practice.yaml \
+            > /tmp/practice-app.yaml
+          helm template medical-chatbot-data infra/helm/medical-chatbot \
+            --values infra/helm/medical-chatbot/values-practice-data.yaml \
+            > /tmp/practice-data.yaml
+
+          # Behavioural parity with the audited production runtime contract.
+          grep -q 'ANSWER_MODEL_ID: "qwen.qwen3-next-80b-a3b"' /tmp/practice-app.yaml
+          grep -q 'ANSWER_PROVIDER: "bedrock-converse"' /tmp/practice-app.yaml
+          grep -q 'EMBEDDING_PROVIDER: "cohere-v4"' /tmp/practice-app.yaml
+          grep -q 'EMBEDDING_DIMENSIONS: "1024"' /tmp/practice-app.yaml
+          grep -q 'EVIDENCE_MINIMUM_SCORE: "0.12"' /tmp/practice-app.yaml
+          grep -q 'RERANK_ENABLED: "true"' /tmp/practice-app.yaml
+          grep -q 'AWS_REGION: "us-east-1"' /tmp/practice-app.yaml
+          grep -q 'checksum/runtime-config:' /tmp/practice-app.yaml
+          grep -q 'host: readytochat.realvuxbaro.me' /tmp/practice-app.yaml
+
+          # The app release must own neither data StatefulSet: PostgreSQL and
+          # Qdrant belong to the data release, so an app-side sync failure or
+          # prune can never delete the corpus or the query history. Only those
+          # two use volumeClaimTemplates — the observability PVCs are the app
+          # release's own and are expected here.
+          #
+          # `set -e` ignores a command whose status is inverted with `!`, so
+          # every must-NOT-contain assertion is written as an explicit exit.
+          refute() {
+            if grep -q "$2" "$1"; then
+              echo "::error::$1 must not contain: $2"
+              exit 1
+            fi
+          }
+
+          refute /tmp/practice-app.yaml 'volumeClaimTemplates'
+          grep -q 'medical-chatbot-data-medical-chatbot-qdrant' /tmp/practice-app.yaml
+
+          # ...and the data release must own nothing else.
+          refute /tmp/practice-data.yaml 'medical-chatbot-data-medical-chatbot-ai-service'
+          refute /tmp/practice-data.yaml 'kind: Ingress'
+          grep -q 'volumeClaimTemplates' /tmp/practice-data.yaml

coordination/ARGOCD_PRODUCTION_MIGRATION_STATE_2026-08-17.md

@@ -0,0 +1,250 @@
+# ArgoCD production migration state — 2026-08-17
+
+## Purpose
+
+Shared handoff for Codex and Claude. This is the current operational reference
+for the personal production-to-ArgoCD migration. Re-check live state before a
+mutation because runtime and Git revisions can change after this snapshot.
+
+## Current verdict
+
+**Latency parity achieved for the measured post-fix scope.** After aligning the
+personal k3s practice environment with production on Qwen and reranking, a
+controlled five-pair test showed practice at 12.194 seconds average and
+production at 12.842 seconds average, with zero errors or timeouts. There is no
+measured systematic k3s latency penalty remaining.
+
+This verdict is specifically about runtime latency parity. It does not mean the
+production migration is ready to cut over: tracked ArgoCD values, immutable
+production images, state restore, DNS rollback and broader soak/regression gates
+remain open.
+
+## Owner intent
+
+- Production must eventually run through ArgoCD.
+- The separate personal k3s/ArgoCD EC2 is the rehearsal environment used to
+  discover migration risk before cutover.
+- The intended cutover is to move `realvuxbaro.me` to the proven k3s workload.
+- Keep the existing Compose production EC2 intact as the DNS-level rollback
+  until the ArgoCD deployment has passed its acceptance window.
+
+## Hard boundary
+
+- Do not access, modify, or push to `git.vinmec.tech`.
+- Do not access or mutate the team's ArgoCD/k3s infrastructure.
+- The active scope is the owner's personal GitHub repository, personal AWS
+  account, Compose production EC2, and personal practice k3s/ArgoCD EC2.
+- No team Gitea or team infrastructure action occurred in the 2026-08-17 Codex
+  session. Reading an old Claude memory that mentioned Gitea did not authorize
+  or cause a connection to it.
+
+## Topology at this snapshot
+
+| Role | Runtime | Address / identity | State |
+| --- | --- | --- | --- |
+| Current production | Docker Compose on EC2 `i-039fc8f6102467a54` (`t3.large`) | `realvuxbaro.me`, `52.0.158.61` | Live; rollback source |
+| Migration rehearsal | k3s + ArgoCD on EC2 `i-035cd1f80f4462455` (`t3.large`) | `readytochat.realvuxbaro.me`, `argocd.realvuxbaro.me` | Live, Synced, Healthy |
+| Source repository | Private GitHub | `BaoVu2k4/vsf-duocthu` | Active source for both personal environments |
+
+The practice instance is currently `t3.large`. Older Claude memory that says it
+is still `t3.medium` is stale.
+
+## Production evidence collected by Codex
+
+Codex added and manually dispatched the read-only workflow
+`.github/workflows/audit-production-runtime.yml`.
+
+- Commit: `2ff65d9` (`Add read-only production runtime audit`).
+- GitHub Actions run: `31993964589`, passed.
+- The commit touched only the new manual audit workflow. It did not match the
+  existing production deploy workflow's path filters and did not restart the
+  Compose stack.
+- Production checkout revision: `df57e6b0806daccb13970dff2aecb7b7dd33eddd`,
+  branch `master`.
+- `ai-service`, `web`, PostgreSQL, and Qdrant containers: running.
+- `/health`: HTTP 200; `/ready`: HTTP 200.
+- Qdrant collection: `duocthu_v1`, 15,100 points, green, cosine, 1,024
+  dimensions.
+- Persistent data mounts: Docker volumes `docker_postgres-data` and
+  `docker_qdrant-data`.
+
+Safe production AI runtime contract:
+
+| Setting | Effective value |
+| --- | --- |
+| `ANSWER_PROVIDER` | `bedrock-converse` |
+| `ANSWER_MODEL_ID` | `qwen.qwen3-next-80b-a3b` |
+| `EMBEDDING_PROVIDER` | `cohere-v4` |
+| `EMBEDDING_DIMENSIONS` | `1024` |
+| `EVIDENCE_MINIMUM_SCORE` | `0.12` |
+| `RERANK_ENABLED` | `true` |
+| `MAX_WALL_CLOCK_MS` | `40000` |
+| `MAX_LLM_CALLS_PER_TURN` | `8` |
+| `AWS_REGION` | `us-east-1` |
+| `QDRANT_COLLECTION` | `duocthu_v1` |
+| `OTEL_ENABLED` | `true` |
+| `OTEL_SAMPLE_RATIO` | `1.0` |
+
+No credential, secret value, patient data, or raw `.env.prod` content was
+printed or stored.
+
+## Practice evidence collected by Codex
+
+The ArgoCD API was read using the gitignored local practice credential file.
+No practice Application or cluster resource was mutated during this audit.
+
+- Application `medical-chatbot-app`: target `master`, compared revision
+  `f9c20a67943e1df3c9e0cb08d01ebaaee48c0470`, Synced, Healthy, automated
+  prune and self-heal enabled.
+- AI image:
+  `ghcr.io/baovu2k4/vsf-duocthu-ai-service:f9c20a67943e1df3c9e0cb08d01ebaaee48c0470`.
+- Application `medical-chatbot-data` separately owns PostgreSQL and Qdrant.
+- Practice AI config explicitly selects `deepseek.v3.2` and does not expose a
+  Helm value for `RERANK_ENABLED`, so the application default is `false`.
+- Practice and production therefore differ in at least two behavior-changing
+  settings: answer model and rerank enablement.
+
+## Corrected conclusions
+
+- Config drift is real: production runs Qwen with reranking enabled; practice
+  runs DeepSeek with reranking disabled.
+- The drift is a strong explanation for practice tail latency and
+  `request_budget_exhausted`, but causality must be confirmed by running the
+  same model/config and an interleaved benchmark.
+- Claude's claim that both environments were on the exact same commit
+  `f9c20a6` was false. Production was observed at `df57e6b`; practice was at
+  `f9c20a6`. The commits between them did not change AI application source, so
+  this correction does not remove the proven configuration drift.
+- Production behavior/data/config is the migration baseline. Docker Compose is
+  not the target architecture; it remains the rollback implementation.
+
+## Runtime parity remediation applied
+
+Codex fixed and deployed the main behavior-changing drift to the personal
+practice environment on 2026-08-17.
+
+- Commit `5a1a600` added Helm mappings for embedding dimensions, evidence
+  threshold, AWS region and rerank enablement. It also added a runtime-config
+  checksum to the AI Deployment Pod template so a ConfigMap change causes an
+  actual rollout instead of leaving the old process alive.
+- `values-prod.yaml` now records the observed production baseline: Qwen
+  `qwen.qwen3-next-80b-a3b`, Cohere v4, 1,024 dimensions, evidence threshold
+  `0.12`, reranking enabled and AWS region `us-east-1`.
+- Helm validation run `31995055293`: one chart linted, zero failed; default and
+  production manifests rendered and asserted.
+- Full CI run `31995055272`: ingestion pytest, AI ruff+pytest, and web
+  lint+build all passed.
+- The personal ArgoCD practice Application accepted the Qwen+rerank baseline
+  and converged at revision `5a1a600`: Synced and Healthy.
+- Effective practice manifest after convergence: Qwen, rerank `true`, Cohere
+  v4, 1,024 dimensions, threshold `0.12`, `us-east-1`, and a non-empty
+  `checksum/runtime-config` Pod annotation.
+- CloudWatch over the post-rollout six-minute window recorded 24 Qwen
+  invocations and no DeepSeek invocation. This confirms the exercised path no
+  longer used DeepSeek; CloudWatch remains account-aggregate rather than
+  environment-labelled.
+
+Small live parity sample through each environment's real `/api/chat` path:
+
+| Scope | Practice | Production |
+| --- | ---: | ---: |
+| Initial smoke | 8.75 s | 8.94 s |
+| Three explicit attribute questions — average | 15.33 s | 11.96 s |
+| Three explicit attribute questions — maximum | 22.84 s | 15.54 s |
+| Errors/timeouts in the three-question sample | 0 | 0 |
+
+All three explicit questions were answerable with grounded evidence and one
+citation in both environments. The initial unaccented overview question was
+answerable on practice but clarified for a missing attribute on production, so
+behavioral determinism still needs a larger test set.
+
+Conclusion: the major model/rerank drift is fixed and the observed 2–3x tail /
+40-second timeout pattern did not reproduce after parity. Practice remained
+about 28% slower on the three-question sample and had a 22.84-second maximum,
+so this is evidence of a material fix, not proof that all residual latency is
+gone. Run a larger interleaved soak before production cutover.
+
+### Residual-latency investigation
+
+The apparent 28% residual above did not survive a controlled follow-up. Five
+additional pairs used the same explicit Zolpidem attribute query, alternated
+which environment ran first, used fresh conversations, and captured each Tempo
+trace.
+
+| Environment | N | Average | Median | Minimum | Maximum | Errors |
+| --- | ---: | ---: | ---: | ---: | ---: | ---: |
+| Practice | 5 | 12.194 s | 12.036 s | 11.876 s | 12.691 s | 0 |
+| Production | 5 | 12.842 s | 12.333 s | 10.803 s | 15.217 s | 0 |
+
+Practice was 0.648 seconds faster on average in this controlled sample. There
+is no remaining systematic practice slowdown demonstrated by the data.
+
+Tempo provider-span averages:
+
+| Operation | Practice | Production |
+| --- | ---: | ---: |
+| Qwen understanding | 5.456 s | 4.795 s |
+| Qwen generation | 5.271 s | 5.793 s |
+| Qwen entailment | 0.711 s | 0.640 s |
+
+The distributions overlap and the faster side changes per call/stage. Output
+length explains the clearest tail: production round 5 returned 1,426
+characters and spent 8.048 seconds in generation; the common 910-character
+outputs spent roughly 4.5–5.4 seconds. Restricting comparison to the common
+910-character outputs gave generation averages of about 5.14 seconds on
+practice and 5.23 seconds on production.
+
+Infrastructure checks also ruled out the earlier throttle hypothesis:
+
+- both instances are `t3.large` in the same `us-east-1d` availability zone,
+  subnet and VPC;
+- practice CPU averaged 15.7% and peaked at 34.87% over the observed 30-minute
+  window;
+- practice retained roughly 599 CPU credits and had zero surplus-credit usage;
+- retrieval remained millisecond-scale, while the varying time was inside
+  Bedrock provider spans.
+
+Final latency diagnosis: the former large gap was caused by the now-fixed
+DeepSeek/rerank configuration drift. The smaller post-fix gap from the first
+three queries was sampling noise from managed Bedrock inference and variable
+generated-output length, not a persistent k3s penalty. No additional
+environment-specific latency fix is justified by the measured state. Reducing
+the absolute approximately 12-second response time would be a separate RAG
+architecture change because a normal turn currently performs understanding,
+generation and entailment as sequential Qwen calls.
+
+## Migration risks currently open
+
+1. The live practice Applications use untracked inline Helm values. This is not
+   yet a complete Git source of truth.
+2. `values-prod.yaml` still uses mutable `latest` image tags; it is not safe for
+   cutover as written even though its behavior settings now match production.
+3. PostgreSQL/Qdrant snapshot, restore, rollback, ingress/TLS/DNS, secrets,
+   resource limits, and failure recovery still need explicit rehearsal gates.
+
+## Workspace ownership and current edits
+
+- Main worktree `D:\VSF-DUOCTHU`, branch `agent/query-history`, contains
+  pre-existing uncommitted Claude/user changes in three Helm files concerning
+  optional AWS static credentials plus untracked slide/material files. Do not
+  stage, overwrite, or discard them accidentally.
+- Codex created clean worktree `D:\VSF-DUOCTHU-codex-argocd`, branch
+  `agent/argocd-prod-migration`, from `origin/master` to isolate migration work.
+- No Helm migration change has been made in that clean worktree yet at this
+  checkpoint.
+- A stale Claude worktree under `.claude/worktrees/agent-a1f5e73fc2ce814e8`
+  contains unrelated, uncommitted table-reconstruction work from 2026-08-05.
+
+## Next safe execution order
+
+1. Move the practice Application's stable non-secret config out of inline Helm
+   values and into a tracked values file; keep only dynamic image tags and
+   secret references outside Git.
+2. Verify the actual running Pod environment, image digest, Qdrant identity,
+   health, readiness and traces through a cluster-level inspection path.
+3. Run a larger interleaved production/practice benchmark and compare
+   p50/p95/p99,
+   timeout rate, answer decisions and citations.
+4. Rehearse state restore and DNS rollback. Do not repoint `realvuxbaro.me`
+   until the gates and rollback path pass.
+

coordination/CLAUDE_CLAIM_2026-08-17.md

@@ -0,0 +1,40 @@
+# Claude claim — 2026-08-17 (afternoon)
+
+Continuing the Codex ArgoCD migration thread at the owner's instruction. Codex's
+own worktree `D:\VSF-DUOCTHU-codex-argocd` was clean at `1684e10` with nothing
+in flight when this claim was taken.
+
+## Owned by Claude in this session
+
+- `infra/helm/medical-chatbot/values-practice.yaml` (new)
+- `infra/helm/medical-chatbot/values-practice-data.yaml` (new)
+- `.github/workflows/helm-chart.yml` (practice render + assertions)
+- The two personal practice ArgoCD Application specs
+  (`medical-chatbot-app`, `medical-chatbot-data`) — moving inline values into
+  the tracked files above.
+
+Worktree: `D:\VSF-DUOCTHU-claude-gitops`, branch `agent/gitops-tracked-values`.
+
+## Not touched
+
+- `git.vinmec.tech`, team ArgoCD/k3s, team repos — hard boundary, unchanged.
+- The Compose production EC2 and `realvuxbaro.me`. `infra/helm/**` is not in
+  `deploy.yml`'s path filters, so pushing this work cannot restart production.
+- The uncommitted AWS-static-credentials Helm diff in the main worktree
+  `D:\VSF-DUOCTHU` (branch `agent/query-history`). Left exactly as found.
+
+## Which migration risk this closes
+
+Open risk 1 in `ARGOCD_PRODUCTION_MIGRATION_STATE_2026-08-17.md`: the live
+practice Applications carried their entire configuration in untracked inline
+`spec.source.helm.values`, so the cluster could drift from the repository with
+no commit recording it — the same class of failure that left practice on
+DeepSeek with reranking off while production ran Qwen with reranking on.
+
+Image tags stay inline deliberately: `.github/scripts/sync_practice_argocd.py`
+regex-rewrites them on every push, so a tag committed to Git would be stale by
+design. Nothing secret was inline, so nothing secret moves.
+
+Equivalence was checked before the Application specs were edited: both tracked
+files parse to structures identical to the live inline values with only the
+`aiService.image` / `web.image` blocks removed.

infra/helm/medical-chatbot/templates/ai-service.yaml

@@ -10,8 +10,12 @@ data:
   QDRANT_URL: {{ default (printf "http://%s-qdrant:6333" (include "medical-chatbot.fullname" .)) .Values.qdrant.url | quote }}
   QDRANT_COLLECTION: {{ .Values.aiService.config.qdrantCollection | quote }}
   EMBEDDING_PROVIDER: {{ .Values.aiService.config.embeddingProvider | quote }}
+  EMBEDDING_DIMENSIONS: {{ .Values.aiService.config.embeddingDimensions | quote }}
+  EVIDENCE_MINIMUM_SCORE: {{ .Values.aiService.config.evidenceMinimumScore | quote }}
+  AWS_REGION: {{ .Values.aiService.config.awsRegion | quote }}
   ANSWER_PROVIDER: {{ .Values.aiService.config.answerProvider | quote }}
   ANSWER_MODEL_ID: {{ .Values.aiService.config.answerModelId | quote }}
+  RERANK_ENABLED: {{ .Values.aiService.config.rerankEnabled | quote }}
   ENTITIES_PATH: {{ .Values.aiService.config.entitiesPath | quote }}
   METRICS_ENABLED: {{ .Values.aiService.config.metricsEnabled | quote }}
   OTEL_ENABLED: {{ and .Values.observability.enabled .Values.aiService.config.otelEnabled | quote }}
@@ -40,6 +44,7 @@ spec:
         app.kubernetes.io/instance: {{ .Release.Name }}
         app.kubernetes.io/component: ai-service
       annotations:
+        checksum/runtime-config: {{ dict "config" .Values.aiService.config "environment" .Values.global.environment "qdrantUrl" .Values.qdrant.url "observabilityEnabled" .Values.observability.enabled | toJson | sha256sum | quote }}
         prometheus.io/scrape: "true"
         prometheus.io/path: /metrics
         prometheus.io/port: "8000"

infra/helm/medical-chatbot/values-practice-data.yaml

@@ -0,0 +1,28 @@
+# k3s practice cluster — the `medical-chatbot-data` release.
+#
+# Companion to values-practice.yaml. The practice cluster deliberately splits
+# the chart into two ArgoCD Applications so that PostgreSQL and Qdrant survive
+# every app rollout, prune and self-heal: only this release owns the
+# PersistentVolumeClaims, so an app-side sync failure can never delete the
+# 15,100-point corpus or the query history.
+#
+# It therefore runs *only* the stateful half — PostgreSQL and Qdrant are left
+# at their chart defaults (`enabled: true`) and everything else is switched off.
+#
+# Kept inline on the Application rather than here: nothing. This release has no
+# image tag that CI rewrites and no secret, so it is fully tracked in Git.
+
+global:
+  environment: k3s-practice
+
+aiService:
+  enabled: false
+
+web:
+  enabled: false
+
+observability:
+  enabled: false
+
+ingress:
+  enabled: false

infra/helm/medical-chatbot/values-practice.yaml

@@ -0,0 +1,72 @@
+# k3s practice cluster — the `medical-chatbot-app` release.
+#
+# This file is the Git source of truth for every stable, non-secret setting of
+# the practice rehearsal environment. It exists because the same values used to
+# live only inside the ArgoCD Application's inline `spec.source.helm.values`,
+# where they were invisible to review, diff and rollback: the cluster could
+# drift from the repository without a single commit recording it. That is how
+# practice ended up on DeepSeek with reranking off while production ran Qwen
+# with reranking on.
+#
+# Two things deliberately stay OUT of this file and remain inline on the
+# Application:
+#
+#   - `aiService.image` / `web.image` — rewritten on every push by
+#     `.github/scripts/sync_practice_argocd.py`, which regex-replaces the tag
+#     in the inline values. A tag committed here would be stale by design.
+#   - anything secret. Nothing in the practice release currently needs one;
+#     Bedrock authenticates through the node's instance role.
+#
+# ArgoCD applies `valueFiles` in order and then the inline values last, so the
+# image block inline still wins over anything the chart defaults set here.
+#
+# The behavioural settings below mirror the production runtime contract audited
+# on 2026-08-17 (see coordination/ARGOCD_PRODUCTION_MIGRATION_STATE_2026-08-17.md).
+# Practice is only useful as a migration rehearsal while they stay in sync.
+
+global:
+  environment: k3s-practice
+  # The GHCR packages are private; without this the Pods fail ImagePullBackOff.
+  imagePullSecrets:
+    - name: ghcr-pull-secret
+
+# PostgreSQL and Qdrant are owned by the separate `medical-chatbot-data`
+# release so that redeploying the app never touches persistent state. This
+# release therefore runs neither, and points at the data release's Services.
+postgres:
+  enabled: false
+
+qdrant:
+  enabled: false
+  url: http://medical-chatbot-data-medical-chatbot-qdrant.medical-chatbot-data.svc.cluster.local:6333
+
+secret:
+  postgresHost: medical-chatbot-data-medical-chatbot-postgres.medical-chatbot-data.svc.cluster.local
+
+aiService:
+  config:
+    embeddingProvider: cohere-v4
+    embeddingDimensions: 1024
+    evidenceMinimumScore: 0.12
+    awsRegion: us-east-1
+    answerProvider: bedrock-converse
+    answerModelId: qwen.qwen3-next-80b-a3b
+    rerankEnabled: true
+
+observability:
+  grafana:
+    # Practice-only convenience: the rehearsal cluster is a throwaway box and
+    # Grafana sits behind the same ingress host. Never set this on production.
+    anonymousAdmin: true
+
+ingress:
+  enabled: true
+  className: traefik
+  host: readytochat.realvuxbaro.me
+  annotations:
+    cert-manager.io/cluster-issuer: letsencrypt-prod
+    traefik.ingress.kubernetes.io/router.entrypoints: websecure
+  tls:
+    - hosts:
+        - readytochat.realvuxbaro.me
+      secretName: readytochat-tls

infra/helm/medical-chatbot/values-prod.yaml

@@ -7,7 +7,12 @@ aiService:
     tag: latest
   config:
     embeddingProvider: cohere-v4
+    embeddingDimensions: 1024
+    evidenceMinimumScore: 0.12
+    awsRegion: us-east-1
     answerProvider: bedrock-converse
+    answerModelId: qwen.qwen3-next-80b-a3b
+    rerankEnabled: true
     otelSampleRatio: 0.25
 
 web:

infra/helm/medical-chatbot/values.yaml

@@ -31,8 +31,12 @@ aiService:
     port: 8000
   config:
     embeddingProvider: disabled
+    embeddingDimensions: 1024
+    evidenceMinimumScore: 0.12
+    awsRegion: us-east-1
     answerProvider: disabled
     answerModelId: deepseek.v3.2
+    rerankEnabled: false
     # The Dockerfile bakes the catalog file in at this path; config.py's own
     # repo-relative default only resolves inside a full monorepo checkout,
     # not this flattened image layout. Matches apps/ai-service/.env.example.