LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows
Python ai-agentsdata-engineeringdocument-aigenerative-aiidpjson-extractionllmmcp-serverocrpdf-extractionprompt-engineeringstructured-output
Add support to run prompt service directly in local (#43)
Hari John Kuriakose197b9c87fe5d
diff --git a/docker/docker-compose.yaml b/docker/docker-compose.yamlindex 0ca81c1ba..8c75c64d7 100644--- a/docker/docker-compose.yaml+++ b/docker/docker-compose.yaml@@ -24,6 +24,9 @@ services:labels:- traefik.enable=true- traefik.http.routers.backend.rule=Host(`frontend.unstract.localhost`) && PathPrefix(`/api/v1`, `/deployment`)+ extra_hosts:+ # "host-gateway" is a special string that translates to host docker0 i/f IP.+ - "host.docker.internal:host-gateway"# Celery execution consumerexecution-consumer:diff --git a/prompt-service/src/unstract/prompt_service/main.py b/prompt-service/src/unstract/prompt_service/main.pyindex 1abe2e8e8..995287160 100644--- a/prompt-service/src/unstract/prompt_service/main.py+++ b/prompt-service/src/unstract/prompt_service/main.py@@ -928,4 +928,4 @@ def extract_variable(if __name__ == "__main__":# Start the server- app.run(port=5003)+ app.run(host="0.0.0.0", port=5003)