AI
by @ops LLM Gateway / Proxy
A lightweight proxy in front of one or more LLM providers, with Redis for response caching and rate limiting — the pattern in front of the pattern, not another RAG pipeline.
Loads this recipe straight into the wizard — every setting stays editable before you generate anything.
The stack
Install
# pulls this exact recipe straight from the Vault $ npx stacktora install llm-gateway --remote $ stacktora sync # generate the files
What you get
More from the Vault
Python + Postgres + Redis: AI Agent Orchestration / Tool-calling Gateway
Manages multi-step agent tool-calls and orchestration state, a worker processing agent steps async — distinct from the LLM Gateway/Proxy recipe: that's caching and rate limiting in front of model calls, this is the orchestration layer coordinating an agent's actual multi-step work.
.NET + Postgres + Redis: Field Sales CRM & Visit Tracking
Sales rep visit logging, territory assignment, and account history — a genuinely common field-sales vertical not yet in the catalog, distinct from the Field Service Technician recipe (that's repair/installation work, not sales visits).
.NET + Postgres + Redis: Fintech-oriented Baseline
A security-conscious starting point for financial and payments-adjacent backends: .NET API, Postgres for ledger/transaction records, and Redis for session and rate-limit state. This is a starting baseline only, not a certified or verified PCI-DSS/regulatory-compliant stack — your own security and compliance review is still required before handling real financial data.
.NET + Postgres + Redis: Insurance Policy Quoting Engine
Rate calculation and quote generation before a policy exists — distinct from Insurance Claims Processing: that handles claims after something happens, this is underwriting math and quote state before a customer ever buys.