Runtime containment kernel for LLM agents. Enforces budget, step, retry, and circuit-breaker limits before the model call.

AI API gateway with cost tracking and budget enforcement. Works with any language.

UnitCause is a real-time cost governance platform for AI agents. It provides budget enforcement, loop detection, cost analytics, and policy management via a web dashboard, REST API, and Python SDK. Set budgets, stop runaway agents, and monitor spending with ease.

CLI bridge for delegating heavy reasoning from Claude Code to Gemini API. Auto-routing, deep thinking tiers, SHA256 caching, daily budget enforcement. Python.

Example showing how to control Anthropic API costs using the shekel Python library

Before/after examples comparing LLM usage with and without shekel budget enforcement

Example showing how to control OpenAI API costs using the shekel Python library

Example showing how to enforce LLM budgets on LangGraph agents using the shekel Python library

Durable LLM agent runtime — crash recovery, budget limits, and permission enforcement built in

Example showing how to enforce LLM budgets on LangChain using the shekel Python library

Your AI agent just burned $200. AgentGuard stops it at $5. Runtime cost guardrails for AI agents — budget enforcement, loop detection, kill switch. Zero dependencies, MIT licensed.