v2.0.2: Production release with role parsing fix and threshold correction

Dockerfile

@@ -4,15 +4,6 @@
 # Build arguments:
 #   APP_UID: User ID for appuser (default: 999)
 #   APP_GID: Group ID for appgroup (default: 999)
-#
-# Build example:
-#   docker build --build-arg APP_UID=1000 --build-arg APP_GID=1000 -t vera-ai .
-#
-# Runtime environment variables:
-#   TZ: Timezone (default: UTC)
-#   APP_UID: User ID (informational)
-#   APP_GID: Group ID (informational)
-#   VERA_LOG_DIR: Debug log directory (default: /app/logs)
 
 # Stage 1: Builder
 FROM python:3.11-slim AS builder
@@ -20,9 +11,7 @@ FROM python:3.11-slim AS builder
 WORKDIR /app
 
 # Install build dependencies
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    build-essential \
-    && rm -rf /var/lib/apt/lists/*
+RUN apt-get update && apt-get install -y --no-install-recommends     build-essential     && rm -rf /var/lib/apt/lists/*
 
 # Copy requirements and install
 COPY requirements.txt .
@@ -38,29 +27,25 @@ ARG APP_UID=999
 ARG APP_GID=999
 
 # Create group and user with specified UID/GID
-RUN groupadd -g ${APP_GID} appgroup && \
-    useradd -u ${APP_UID} -g appgroup -r -m -s /bin/bash appuser
+RUN groupadd -g ${APP_GID} appgroup &&     useradd -u ${APP_UID} -g appgroup -r -m -s /bin/bash appuser
 
 # Copy installed packages from builder
 COPY --from=builder /root/.local /home/appuser/.local
 ENV PATH=/home/appuser/.local/bin:$PATH
 
 # Create directories for mounted volumes
-RUN mkdir -p /app/config /app/prompts /app/static /app/logs && \
-    chown -R ${APP_UID}:${APP_GID} /app
+RUN mkdir -p /app/config /app/prompts /app/logs &&     chown -R ${APP_UID}:${APP_GID} /app
 
 # Copy application code
 COPY app/ ./app/
 
 # Copy default config and prompts (can be overridden by volume mounts)
-COPY config.toml /app/config/config.toml
-COPY static/curator_prompt.md /app/prompts/curator_prompt.md
-COPY static/systemprompt.md /app/prompts/systemprompt.md
+COPY config/config.toml /app/config/config.toml
+COPY prompts/curator_prompt.md /app/prompts/curator_prompt.md
+COPY prompts/systemprompt.md /app/prompts/systemprompt.md
 
-# Create symlinks for backward compatibility
-RUN ln -sf /app/config/config.toml /app/config.toml && \
-    ln -sf /app/prompts/curator_prompt.md /app/static/curator_prompt.md && \
-    ln -sf /app/prompts/systemprompt.md /app/static/systemprompt.md
+# Create symlink for config backward compatibility
+RUN ln -sf /app/config/config.toml /app/config.toml
 
 # Set ownership
 RUN chown -R ${APP_UID}:${APP_GID} /app && chmod -R u+rw /app
@@ -70,11 +55,10 @@ ENV TZ=UTC
 
 EXPOSE 11434
 
-# Health check using Python (no curl needed in slim image)
-HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
-    CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:11434/')" || exit 1
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3     CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:11434/')" || exit 1
 
 # Switch to non-root user
 USER appuser
 
-CMD ["python", "-m", "uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "11434"]"
+ENTRYPOINT ["python", "-m", "uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "11434"]

config/config.toml

@@ -2,18 +2,15 @@
 ollama_host = "http://10.0.0.10:11434"
 qdrant_host = "http://10.0.0.22:6333"
 qdrant_collection = "memories"
-embedding_model = "snowflake-arctic-embed2"
+embedding_model = "mxbai-embed-large"
 debug = false
 
 [layers]
-# Note: system_token_budget removed - system prompt is never truncated
 semantic_token_budget = 25000
 context_token_budget = 22000
 semantic_search_turns = 2
-semantic_score_threshold = 0.6
+semantic_score_threshold = 0.3
 
 [curator]
-# Daily curation: processes recent 24h of raw memories
-# Monthly mode is detected automatically by curator_prompt.md (day 01)
 run_time = "02:00"
 curator_model = "gpt-oss:120b"

prompts/systemprompt.md

@@ -1,10 +1 @@
-You have persistent memory across all conversations with this user.
 
-**Important:** The latter portion of your conversation context contains memories retrieved from a vector database. These are curated summaries of past conversations, not live chat history.
-
-Use these memories to:
-- Reference previous decisions and preferences
-- Draw on relevant past discussions
-- Provide personalized, context-aware responses
-
-If memories seem outdated or conflicting, ask for clarification.