Founder, Sovereign Logic | Oakland, CA
📧 docspoc101@gmail.com
I build tools that prevent AI agents from burning money and breaking production.
🔗 SHACKLE — AI Agent Circuit Breaker
Pre-execution guardrails for CrewAI, LangGraph, and AutoGen.
Stop runaway costs, infinite loops, and production failures before they happen.
from shackle import Guard
@Guard(budget=1.00, max_repeat_calls=3, timeout_seconds=300)
def protected_agent():
return agent.run() # Safe, monitored, interruptibleThe Problem: AI agents fail silently. They loop. They burn budgets. They destroy infrastructure.
The Solution: Runtime circuit breakers that enforce hard limits at the interpreter level.
📖 Documentation | 🐙 GitHub | 🌐 Website
- Production agent safety — Making AI agents enterprise-ready
- Cost control systems — Budget enforcement for LLM-powered workflows
- Human-in-the-loop tooling — Interactive control for supervised deployments
Building something with AI agents? Running into cost overruns or infinite loops?
I offer architecture audits and implementation consulting.
📧 docspoc101@gmail.com
💬 Available for consulting and collaboration
Built SHACKLE after a $400 loop. Now helping others avoid $106K+ failures.