VERSION 0.4.1

STEERPLANE

Runtime guardrails for autonomous AI agents. Observability tells you your agent failed — SteerPlane stops it before it does. Cost circuit-breakers, deterministic loop termination, and kill/alert enforcement that sit directly in the execution path. One line of code. Zero LLM-judge latency.

$ pip install steerplane$ npm install steerplane

AI agents are powerful. And dangerous.

Autonomous agents call APIs, run code, and make real-world decisions. Without guardrails, a single misconfigured agent can spin into an infinite loop, burn thousands of dollars in API credits overnight, or take a destructive action — and today's observability tools can only tell you about it after the fact. SteerPlane is built to intervene in real time, not just report.

20
Patent claims (IN 202641071111 A1)
4
Framework integrations
25+
LLM models priced
<1ms
Guard-check latency

Guardrails that actually stop the run

Every capability is production-ready and deterministic. Use the zero-code gateway, the SDK decorator, or native framework integrations — and get the same enforcement pipeline across all of them.

Loop Termination

A deterministic O(W²) sliding-window detector spots single-action, alternating, and multi-step repeating patterns in sub-millisecond time — then hard-terminates the run before it drains your budget on redundant calls. No LLM judge, no probabilistic guessing.

Cost Circuit-Breaker

Per-session USD ceilings across 25+ models with built-in pricing. On streaming responses, SteerPlane parses each chunk and severs the connection mid-stream the instant the ceiling is hit — overshoot is bounded to a single step.

Kill / Alert Enforcement

Not a binary kill-or-nothing switch. Kill mode terminates instantly on violation; alert mode pauses, notifies, and waits for human approval — then resumes with extended limits. Loops and policy breaches always hard-terminate, as a non-overridable invariant.

Policy Engine

A hierarchical deny → allow → rate-limit → approval pipeline with glob-pattern matching. Block destructive actions like delete_* and drop_*, rate-limit noisy calls, and gate sensitive operations like refund_* behind human approval — before any LLM call is made.

Zero-Code Gateway

An OpenAI-compatible proxy that sits between your agent and the provider. Change only the base_url — every call is monitored, policy-checked, and cost-tracked. Your real provider key stays server-side and never touches the agent process.

Framework-Agnostic

Drop-in integrations for LangChain, CrewAI, AutoGen, and the OpenAI Agents SDK — plus a Python decorator and TypeScript SDK. SteerPlane registers as the native callback/hook and routes every call through one guardrail pipeline, with your agent code unchanged.

Real-Time Dashboard

A live console that visualizes runs, execution timelines, cost breakdowns, policy violations, and pending approvals — with auto-refresh. Manage API keys and policies, and watch fleet-wide agent health at a glance.

Fault-Tolerant by Design

If the control plane goes offline, the SDK keeps enforcing loop detection, cost, and policy locally — and alert mode automatically degrades to kill mode. No agent ever runs without at least one active enforcement path.

Six steps. Under five minutes.

1

Install

pip install steerplane — or npm install steerplane. One dependency, zero config to start.

2

Wrap or Proxy

Add the @guard decorator, or point your OpenAI client's base_url at the SteerPlane gateway. No agent rewrite.

3

Set Limits

Declare cost ceilings, step limits, denied actions, rate limits, and approval gates — in code or a .steerplane.yml.

4

Agent Runs

Your agent executes normally. SteerPlane intercepts every LLM call and action in the path.

5

Guards Fire

Loops terminate, cost ceilings break the circuit, policies block destructive actions, and alert-mode pauses for approval — in real time.

6

You See Everything

Every run, step, cost, and violation streams to the live dashboard for debugging and audit.

Two integration paths. Total protection.

Wrap a function with the decorator for deep control, or point your OpenAI client at the gateway for zero-code enforcement. Either way you get cost limits, loop detection, policy evaluation, and full telemetry out of the box.

agent.py
Python 3.10+
# ── Option 1: Decorator (deep control) ──────────
from steerplane import guard

@guard(
    agent_name="support_bot",
    max_cost_usd=10.00,
    max_steps=50,
    denied_actions=["delete_*", "drop_*"],
    enforcement="alert",      # pause + ask a human near the limit
)
def run_support_agent():
    agent.run()               # your code, unchanged

# ── Option 2: Gateway (zero code changes) ───────
import openai

client = openai.OpenAI(
    base_url="http://localhost:8000/gateway/v1",
    api_key="sk_sp_...",      # your SteerPlane key
    default_headers={"X-LLM-API-Key": "sk-..."},  # injected server-side
)
# Every call is now monitored, policy-checked, and cost-capped.

A layered control plane

Agent calls are captured by the SDK or gateway, evaluated by the policy engine, loop detector, and cost tracker, then enforced by the kill/alert controller — with every decision persisted and streamed to the dashboard.

AI Agent
SDK / Gateway
Policy Engine
Loop Detector
Cost Tracker
Enforcement
REST API
Dashboard

Built on proven foundations

Python SDK
Python 3.10+
@guard decorator, CLI, .steerplane.yml, LangChain / CrewAI / AutoGen / OpenAI Agents SDK integrations. Published on PyPI.
TypeScript SDK
Node.js 18+
guard() wrapper, async-ready. Published on npm.
Gateway
FastAPI + HTTPX
OpenAI-compatible SSE streaming proxy with mid-stream cost kill and 25+ model pricing.
Control Plane
FastAPI + SQLAlchemy
Runs, steps, policies, approvals, telemetry. Auto OpenAPI docs, Alembic migrations.
Database
PostgreSQL / SQLite
SQLite for dev, PostgreSQL for production. Full run + approval persistence.
Dashboard
Next.js 16 + React 19
Framer Motion, auto-refresh, standalone Docker output.

Ready to secure your agents?

SteerPlane is open-source and free to start. Add one decorator or point your client at the gateway — and start stopping incidents in under five minutes.

SteerPlane v0.4.1 — Runtime guardrails for autonomous AI agents.

steerplane.com · Patent Pending — IN 202641071111 A1