Developer Tools●●Solid
RunVeto – A Simple Kill Switch for Autonomous AI Agents
Agent cost killswitch solves a real pain, but monitoring infra is crowded.
Solve My ProblemShip It
JDPatel1729
103mo ago
Back to browse
GitHub Repository
Deterministic cost / loop / time budgets · full observability · crash-resumable runs · human-approval gates · a memory you own. Self-hosted. Your keys. No telemetry. Point it at your existing agents - one env var.
15 starsGo
RiskKernel – a kill switch and budgets for runaway AI agents
by prashar32·Jun 8, 2026·1 point·0 comments
AI Analysis
●●SolidBig BrainShip It
Deterministic Go governor halts runaway agents at budget with HTTP 402.
Strengths
- •Crash-resumable checkpoints persist budget state across kill -9 and daemon restarts.
- •Framework-agnostic approval gates pause side-effecting tool calls for human review.
- •OpenTelemetry GenAI spans emit to your existing observability backend.
Weaknesses
- •Agent governance is emerging space with multiple competing approaches being tried.
- •Requires self-hosting and one env var integration per agent framework.
Category
Target Audience
Developers running production AI agents who need guardrails
Similar To
Langfuse · Portkey · Guardrails AI
Similar Projects
AI/ML●●●Banger
Varpulis – Real-time behavioral guardrails for AI agents
NFA + ZDD pattern matching catches agent failure modes LangSmith and Guardrails AI miss.
Big BrainZero to OneSolve My Problem
cpo75
102mo ago
Developer Tools●●Solid
ClawSight – Lightweight monitoring and kill switches for AI agents
One-line setup, real problem: $100 bill with zero visibility into what a Telegram agent was doing.
Solve My ProblemShip It
upsidepotential
103mo ago
Developer Tools●●Solid
Preventing runaway LLM agents (enforcement layer)
VERONICA puts an enforcement shim between your agent and the model so you can halt costly spirals before a request hits the provider — it natively exposes hard budget enforcement, circuit breakers, retry containment and degradation levels. The README + runnable runaway-loop demo make the failure mode concrete and the API (BudgetEnforcer, RuntimeContext, BudgetExceeded) is small and practical. I'd like to see richer observability/adapter docs for common agent frameworks, but as an enforcement-first primitive this is a clever, useful tool.
Niche GemBig Brain
amabito
123mo ago
Developer Tools●●●Banger
Cap the spending of your agents
Provider-enforced caps mean rogue agents literally cannot exceed your budget.
Solve My ProblemBig BrainCozy
kristopolous
202mo ago
AI/ML●●Solid
Nerve – self hosted runtime for AI agents
Self-hosted agent runtime with persistent memory and personality modes via SOUL.md files.
SlickSolve My Problem
animetyan
6413d ago