Dockerfile

FROM ghcr.io/vectorize-io/hindsight:latest

# HuggingFace Spaces requires port 7860
ENV HINDSIGHT_API_HOST=0.0.0.0
ENV HINDSIGHT_API_PORT=7860
ENV HINDSIGHT_API_LOG_LEVEL=info
ENV HINDSIGHT_API_MCP_ENABLED=true
ENV HINDSIGHT_API_RETAIN_EXTRACTION_MODE=concise
ENV HINDSIGHT_API_RETAIN_EXTRACT_CAUSAL_LINKS=true
ENV HINDSIGHT_API_RERANKER_PROVIDER=flashrank
ENV HINDSIGHT_API_TENANT_EXTENSION=hindsight_api.extensions.builtin.tenant:ApiKeyTenantExtension

# LLM: Google Gemini (set via HF Space secrets)
# HINDSIGHT_API_LLM_PROVIDER=gemini
# HINDSIGHT_API_LLM_API_KEY=<gemini-api-key>
# HINDSIGHT_API_LLM_MODEL=gemini-2.5-flash

# Data directory (ephemeral on HF free tier — backup handles persistence)
ENV HINDSIGHT_DATA_DIR=/data/hindsight

# Backup configuration (set via HF Space secrets)
# HF_TOKEN=hf_xxxxx
# HF_BACKUP_REPO=Arnwald84/atum-hindsight-backup
ENV BACKUP_INTERVAL_SECONDS=21600

# Install backup dependency and create data directory
USER root
RUN pip install --no-cache-dir huggingface_hub>=0.27.0
RUN mkdir -p /data/hindsight && chown hindsight:hindsight /data/hindsight

# Copy backup/restore scripts
COPY scripts/ /opt/backup/
RUN chmod +x /opt/backup/entrypoint.sh

# Switch back to original user
USER hindsight

EXPOSE 7860

HEALTHCHECK --interval=30s --timeout=10s --retries=5 --start-period=300s \
    CMD curl -f http://localhost:7860/health || exit 1

# Wrap original CMD with backup/restore entrypoint
ENTRYPOINT ["/opt/backup/entrypoint.sh"]
# CMD ["--model-id", "sentence-transformers/all-MiniLM-L6-v2", "--port", "7860"]