v0.1.0 is live— LoopStorm Guard is now open source. Star us on GitHub →

Runtime Guardrails for AI Agents

Intercept every tool call. Enforce policies. Cap budgets. Detect loops. Write tamper-evident audit logs. All before your agent touches production.

No credit card required · 2-minute setup · MIT licensed

< 1ms
P99 Latency
67+
Unit Tests
0
Network Deps
5
Control Stages

Works with every AI framework that makes tool calls

OpenAIAnthropicLangChainLlamaIndexCrewAIAutoGenGoogle GeminiMistral

See It Work

Watch LoopStorm Intercept in Real Time

Every tool call is evaluated against your policy. Allowed, denied, or killed — with a full audit trail.

loopstorm — enforcement log

How It Works

Five Stages of Control

Stages 1 – 4 are fully deterministic. No AI in the critical path. Stage 5 is advisory only — human approval required.

Stage 1Deterministic

Prevent

Policy enforcement at the call boundary.

Binary allow or deny. First-match-wins rule evaluation. YAML-defined. P99 < 1ms latency.

Stage 2Deterministic

Detect

Loop-detection heuristics within a run.

Same call fingerprint? Error loops? Stuck retry cycles? Detected immediately — no ML required.

Stage 3Deterministic

Recover

Cooldown + corrective context injection.

One structured chance for the agent to self-correct. Bounded: if it fails, escalation is automatic.

Stage 4Deterministic

Contain

Safe termination with evidence preservation.

Clean shutdown, not a process kill. Audit log sealed. Evidence preserved for human review.

Stage 5AI-Assisted

Adapt

AI Supervisor on the observation plane.

Interprets patterns. Proposes policy updates. Escalates to humans. Never in the enforcement critical path.

Advisory only · Human approval required · Self-guarded

Capabilities

Everything You Need

A complete safety layer between your AI agents and production systems.

Policy Rules

YAML-defined rules. Deny cloud metadata, block unauthorized tools, restrict URLs. First match wins.

Budget Caps

Hard limits on cost (USD), tokens, and call count per run. Fail-closed when exceeded.

Loop Detection

Fingerprint-based heuristics catch retry loops, error loops, and stuck agents automatically.

Hash-Chain Audit Log

Tamper-evident JSONL log with SHA-256 hash chain. CLI-verifiable. Any tampering is detectable.

Mode 0 — Air-Gapped

Runs entirely local. No account. No network. No telemetry. pip install and go.

Open Core

Engine + CLI + shims are MIT. Use them anywhere. Control plane UI is AGPL-3.0.

Quick Start

Integrate in Under 2 Minutes

Add LoopStorm Guard to your agent in Python or TypeScript. Define your policy in YAML.

agent.pyPython
from loopstorm import guard

@guard(policy="loopstorm.yaml")
def my_agent():
    # Every tool call is now intercepted,
    # policy-checked, budget-tracked, and logged
    result = call_tool("web_search", {"query": "..."})
    return result
agent.tsTypeScript
import { guard } from "loopstorm-ts";

const run = guard({ policy: "loopstorm.yaml" });

// Wrap your agent's tool calls
const result = await run.wrap("web_search", {
  query: "..."
});
loopstorm.yamlPolicy
# loopstorm.yaml
agent_role: my-agent
rules:
  - name: deny-cloud-metadata
    action: deny
    tool_pattern: "http.*"
    conditions:
      - field: url
        operator: matches
        pattern: "169.254.169.254.*"
budget:
  cost_usd:
    hard: 5.00
  calls:
    hard: 100

Pricing

Free to Start. Built to Scale.

The open-source engine is free forever. The cloud dashboard gives your team visibility.

MIT

Open Source

$0 / forever

Full enforcement engine. No limits. No account. No network dependency.

  • Rust enforcement engine
  • Python + TypeScript shims
  • CLI (validate, verify, replay)
  • Policy rules + budget caps
  • Loop detection
  • Hash-chain audit logs
  • Air-gapped (Mode 0)
Get Started
RECOMMENDED
CLOUD

Pro

$49 / month

Everything in Open Source, plus a hosted dashboard for team visibility.

  • Everything in Open Source
  • Web dashboard
  • Team management (5 seats)
  • Cross-run analytics
  • Event ingest API
  • 7-day log retention
  • Email support
Start Free Trial

7-day free trial · No credit card required

ENTERPRISE

Enterprise

Custom

Self-hosted or managed. Full data sovereignty. AI Supervisor on the observation plane.

  • Everything in Pro
  • Unlimited seats
  • Self-hosted option (Helm/Docker)
  • SSO / SAML
  • 90-day log retention
  • AI Supervisor (advisory)
  • SLA + priority support
Contact Sales

Open Source

Proudly Open Source

We believe the best safety tools are built in the open. Inspect every line. Deploy anywhere. Contribute back.

Contribute

Engine (Rust), shims (Python/TS), and CLI are MIT licensed. PRs welcome.

Self-Host

Deploy the full control plane on your infrastructure. Docker, Kubernetes, or bare metal.

Verify Everything

Schema hashes in VERIFY.md. Audit logs are CLI-verifiable. Reproducible builds.

FAQ

Frequently Asked Questions

Less than 1ms P99 per tool call. The enforcement engine is written in Rust, runs locally in-process via IPC (Unix domain socket), and does zero network I/O in Mode 0. Policy evaluation is a single-pass first-match lookup — no ML inference, no API calls. Your agent won't notice it's there.

Under 2 minutes. For Python: pip install loopstorm-py, add @guard(policy="loopstorm.yaml") to your agent, and define your rules. For TypeScript: bun add loopstorm-ts with a similar one-line wrapper. No infrastructure, no account, no config server.

Yes — that's Mode 0, and it's the default. The engine, CLI, and shims run entirely locally. No telemetry, no license server, no phone-home. Audit logs stay on disk. You can air-gap it completely. The cloud dashboard (Mode 2) is optional, for teams that want cross-run analytics.

LoopStorm intercepts at the tool-call level, so it works with any framework that makes tool calls: LangChain, LlamaIndex, CrewAI, AutoGen, custom agents, or raw OpenAI/Anthropic SDK usage. Python and TypeScript shims are included. The Rust engine speaks a simple JSON-over-IPC protocol if you need to integrate from another language.

Most guardrails libraries validate LLM output text (prompt injection, toxicity). LoopStorm operates at the tool-call boundary — it controls what your agent can do, not what it says. Policy rules, budget caps, and loop detection are deterministic (no AI in the critical path). It's an enforcement layer, not a content filter.

Yes. The MIT-licensed engine has 67+ unit tests, 11 integration tests, and 4 end-to-end case studies covering budget exhaustion, loop detection, policy deny, and escalation. The audit log uses SHA-256 hash chains that are CLI-verifiable. It's the same engine that powers the cloud version — there's no "lite" edition.

Stop Your Agents Before They Stop You

LoopStorm Guard is open source, free, and installs in under a minute. No account required. No network dependency. Just safety.

pip install loopstorm-py · MIT licensed