switch-openconcho-to-fork

- Update Dockerfile to clone from code.lazyworkhorse.net/Hermes/honcho.git
  (uses build arg HONCHO_REPO, can be overridden at build time)
- Add config.toml volume mount from HoardingCow persistent path
- Use named volume honcho_data instead of host bind mount
- Declare honcho_data as external volume in top-level volumes section

ai/compose.yml

@@ -1,4 +1,3 @@
-version: "3.8"
 services:
 
   # webui:
@@ -162,23 +161,31 @@ services:
       - "26"
 
   # --- Honcho + OpenConcho combiné: API + Web UI nginx/FastAPI ---
-  honcho-ui:
-    build: ./honcho
-    container_name: honcho-ui
+  honcho:
+    build:
+      context: ./honcho
+      ssh:
+        - default
+    container_name: honcho
     restart: unless-stopped
     environment:
       - DB_CONNECTION_URI=postgresql+psycopg://honcho:honcho_pass@honcho-db:5432/honcho
       - CACHE_URL=redis://honcho-redis:6379/0
       - CACHE_ENABLED=true
-      - EMBEDDING_VECTOR_DIMENSIONS=1536
-      - VECTOR_STORE_DIMENSIONS=1536
+      - EMBEDDING_VECTOR_DIMENSIONS=1024
+      - AUTH_USE_AUTH=true
+      - AUTH_JWT_SECRET=${HONCHO_AUTH_JWT_SECRET}
+      # Needed by deriver/dream to make LLM calls (api_key_env = "HONCHO_OPENAI_API_KEY" in config.toml)
+      - HONCHO_OPENAI_API_KEY=${HONCHO_OPENAI_API_KEY}
     volumes:
-      - /mnt/HoardingCow_docker_data/Honcho/data:/app/data
+      - honcho_data:/app/data
+      - /mnt/HoardingCow_docker_data/Honcho/config.toml:/app/config.toml:ro
     networks:
       - ai_backend
       - ai_net
     labels:
       - "traefik.enable=true"
+      - "traefik.docker.network=ai_net"
 
       # Router for HTTP + redirect to HTTPS
       - "traefik.http.routers.honcho-http.rule=Host(`honcho.lazyworkhorse.net`)"
@@ -246,6 +253,11 @@ networks:
   ai_backend:
     driver: bridge
     name: ai_backend
+
+volumes:
+  honcho_data:
+    external: true
+    name: honcho_data
     
   # llama_cpp_devstral:
   #   image: ghcr.io/ggml-org/llama.cpp:server-rocm

ai/honcho/Dockerfile

@@ -2,14 +2,15 @@
 FROM python:3.13-slim-bookworm AS honcho-builder
 
 RUN apt-get update && \
-    apt-get install -y --no-install-recommends git && \
+    apt-get install -y --no-install-recommends git openssh-client && \
     rm -rf /var/lib/apt/lists/*
 
 COPY --from=ghcr.io/astral-sh/uv:0.9.24 /uv /bin/uv
 
-ARG HONCHO_REPO=https://github.com/plastic-labs/honcho
+ARG HONCHO_REPO=ssh://git@code.lazyworkhorse.net:2222/Hermes/honcho.git
 ARG HONCHO_REF=main
-RUN git clone --depth 1 --branch ${HONCHO_REF} ${HONCHO_REPO} /app
+RUN mkdir -p -m 0700 ~/.ssh && ssh-keyscan -p 2222 code.lazyworkhorse.net >> ~/.ssh/known_hosts 2>/dev/null
+RUN --mount=type=ssh git clone --depth 1 --branch ${HONCHO_REF} ${HONCHO_REPO} /app
 
 WORKDIR /app
 
@@ -29,8 +30,8 @@ RUN corepack enable && corepack prepare pnpm@latest --activate
 WORKDIR /app
 RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*
 
-ARG OPENCONCHO_SHA=e490d911fcb27ee193558fd9a28856cde2057665
-RUN git clone --depth 1 https://github.com/offendingcommit/openconcho.git /app && \
+ARG OPENCONCHO_SHA=3b5c3293fc18d768dbe85285264a8d66c896bd81
+RUN --mount=type=ssh git clone --depth 1 ssh://git@code.lazyworkhorse.net:2222/gortium/openconcho.git /app && \
     git -C /app fetch --depth 1 origin ${OPENCONCHO_SHA} && \
     git -C /app checkout ${OPENCONCHO_SHA}
 
@@ -71,4 +72,4 @@ COPY honcho-nginx.conf /etc/nginx/conf.d/default.conf
 
 EXPOSE 80
 
-CMD ["sh", "-c", "nginx -g 'daemon off;' & fastapi run --host 127.0.0.1 --port 8000 src/main.py"]
+CMD ["bash", "-c", "nginx -g 'daemon off;' & fastapi run --host 127.0.0.1 --port 8000 src/main.py & python3 -m src.deriver & wait -n"]

ai/honcho/config.toml

@@ -29,17 +29,17 @@ URL = "redis://honcho-redis:6379/0"
 [llm]
 DEFAULT_MAX_TOKENS = 4096
 
-# Embeddings via Ollama (nomic-embed-text recommended on this system)
+# Embeddings via Ollama — bge-m3 provides 1024-dim
 [embedding]
-VECTOR_DIMENSIONS = 768
+VECTOR_DIMENSIONS = 1024
 MAX_INPUT_TOKENS = 8192
 
 [embedding.model_config]
 transport = "openai"
-model = "nomic-embed-text"
-base_url = "http://ollama:11434/v1"
+model = "bge-m3"
+overrides = {base_url = "http://ollama:11434/v1", api_key = "ollama"}
 
-# --- Deriver (user representation builder) ---
+# --- Deriver ---
 [deriver]
 ENABLED = true
 WORKERS = 1
@@ -47,9 +47,9 @@ POLLING_SLEEP_INTERVAL_SECONDS = 5.0
 FLUSH_ENABLED = true
 
 [deriver.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 # --- Dialectic ---
 [dialectic]
@@ -60,37 +60,37 @@ SESSION_HISTORY_MAX_TOKENS = 8192
 MAX_TOOL_ITERATIONS = 1
 MAX_OUTPUT_TOKENS = 512
 [dialectic.levels.minimal.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 [dialectic.levels.low]
 MAX_TOOL_ITERATIONS = 3
 [dialectic.levels.low.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 [dialectic.levels.medium]
 MAX_TOOL_ITERATIONS = 2
 [dialectic.levels.medium.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 [dialectic.levels.high]
 MAX_TOOL_ITERATIONS = 4
 [dialectic.levels.high.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 [dialectic.levels.max]
 MAX_TOOL_ITERATIONS = 10
 [dialectic.levels.max.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 # --- Summary ---
 [summary]
@@ -99,13 +99,28 @@ MESSAGES_PER_SHORT_SUMMARY = 20
 MESSAGES_PER_LONG_SUMMARY = 60
 
 [summary.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
 
 # --- Dream ---
 [dream]
-ENABLED = false
+ENABLED = true
+
+[dream.model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
+transport = "openai"
+model = "deepseek-v4-flash"
+
+[dream.deduction_model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
+transport = "openai"
+model = "deepseek-v4-flash"
+
+[dream.induction_model_config]
+overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
+transport = "openai"
+model = "deepseek-v4-flash"
 
 # --- Peer Card ---
 [peer_card]
@@ -114,4 +129,4 @@ ENABLED = true
 # --- Vector Store ---
 [vector_store]
 TYPE = "pgvector"
-DIMENSIONS = 768
+# DIMENSIONS is deprecated — EMBEDDING.VECTOR_DIMENSIONS is authoritative