Deploy and Host Prism Agent Stack on Railway

Prism Agent Stack is a deployable multi-service workspace for running Codex with shared memory, Discord chat, voice meeting capture, change request workflows, and repo-aware automation. It combines a lightweight admin site, Codex runtime, Prism Memory, Discord adapter, and background cron services into one Railway template.

About Hosting Prism Agent Stack

Hosting Prism Agent Stack on Railway creates a full operational workspace for an AI-assisted community or team. The stack deploys multiple connected services, persistent volumes, internal networking, and optional public endpoints for chat, memory artifacts, and admin access. After deploy, the main setup tasks are configuring environment variables, authenticating Codex in the runtime, optionally connecting Discord and voice transcription, and registering target repositories for change requests. Railway handles service orchestration, redeploys, and persistent storage, while Prism Memory keeps shared context and artifacts available across sessions.

Common Use Cases

• Run a Codex-powered change request board that can branch, edit, and push updates to target GitHub repositories.
• Add a Discord bot that can answer questions, create threaded agent sessions, and capture voice meeting transcripts and summaries.
• Maintain shared project memory and searchable knowledge so agents can retrieve context, artifacts, and handbook content across sessions.

Dependencies for Prism Agent Stack Hosting

• A GitHub account and at least one repository for template source sync or change-request targets.
• Optional third-party credentials for enabled integrations, such as Discord bot access and a voice transcription provider.

Deployment Dependencies

• GitHub: https://github.com/
• Discord Developer Portal: https://discord.com/developers/applications
• Railway Docs: https://docs.railway.com/
• OpenAI Codex CLI auth flow: https://github.com/openai/codex
• Venice AI audio transcription API (optional): https://docs.venice.ai/

Implementation Details

Prism Agent Stack deploys multiple Railway services with shared internal networking and persistent volume mounts. A typical setup includes:

• site for admin UI and change request management
• api for application state and orchestration
• codex-runtime for Codex execution and Git operations
• prism-memory for shared memory, knowledge, and artifact storage
• discord-adapter for Discord chat and voice flows
• memory-cron, knowledge-cron, and discord-sync-cron for scheduled background work

Example internal service wiring:

APP_API_BASE_URL="http://${{api.RAILWAY_PRIVATE_DOMAIN}}:${{api.PORT}}"
CODEX_RUNTIME_BASE_URL="http://${{codex-runtime.RAILWAY_PRIVATE_DOMAIN}}:${{codex-runtime.PORT}}"
PRISM_MEMORY_BASE_URL="http://${{prism-memory.RAILWAY_PRIVATE_DOMAIN}}:${{prism-memory.PORT}}"

Example public artifact or API wiring:

NEXT_PUBLIC_API_BASE_URL="https://${{api.RAILWAY_PUBLIC_DOMAIN}}"
PRISM_API_BASE="https://${{prism-memory.RAILWAY_PUBLIC_DOMAIN}}"

## Why Deploy Prism Agent Stack on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you
don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Prism Agent Stack on Railway, you are one step closer to supporting a complete full-stack
application with minimal burden. Host your servers, databases, AI agents, and more on Railway.