Runtime Control Plane for AI

Govern AI Agents in Real-Time

A control plane that sits between your AI agents and the real world—deciding what actions are allowed, recording everything, and giving you complete visibility into what happened, why, and what it cost.

Free Tier Available
No Credit Card Required
Python SDK
control-plane.log
[+] Run started — user: alice
→ Step: MODEL_CALL
→ Model: gpt-4o
→ Decision: ALLOW
→ Cost: $0.0023
[!] Budget check: user daily limit
→ Spent today: $50.02 / $50.00
[x] Step blocked
→ Reason: USER_DAILY_BUDGET_EXCEEDED

Everything You Need to Control AI Agents

A complete control plane for governing, monitoring, and tracking AI agents in production.

Runtime Guardrails

Kill switch, daily budgets, and concurrent run limits enforce boundaries at every step. Automatically deny actions that exceed your configured limits.

Complete Observability

Every model call, tool use, and decision is recorded with full context. See exactly what your agents did and why.

Cost Tracking & Budgets

Track costs per run, per user, and per day. Set workspace and per-user daily budgets to prevent overspending.

Per-User Limits

Set daily budgets per SDK user. Users transition from normal to warning to blocked as they approach limits.

Concurrency Control

Limit how many agent runs execute simultaneously. Prevent resource exhaustion and manage capacity.

Dashboard & Analytics

Monitor runs in real-time, view step timelines, track usage by model and user, and manage workspace settings.

How It Works

The control plane sits between your AI agent and the outside world

🤖

AI Agent

Wants to take an action

Control Plane

Evaluates guardrails

🌍

Real World

APIs, Models, Tools

Is this allowed?

Guardrails evaluate in real-time to decide if an action should proceed. Actions that exceed your configured limits are denied automatically.

What happened?

Every action is logged with full context. Trace decisions, debug issues, and audit agent behavior with complete visibility.

What did it cost?

Track costs across models, APIs, and tools. Set budgets, monitor spending, and prevent unexpected charges.

Integrate in Minutes

Wrap your agent code with the SDK. Every model call and tool use is governed and recorded.

agent.py
from pikarc import AsyncPikarc, PikarcBlockedError

guard = AsyncPikarc(
    api_key="lg_xxxx_yyyy",
    base_url="https://api.pikarc.dev",
)

try:
    async with guard.run(user_id="alice") as run:
        response = await run.model_call(
            fn=lambda: openai.chat.completions.create(
                model="gpt-4o",
                messages=[{"role": "user", "content": query}],
            ),
            model="gpt-4o",
            token_extractor=lambda r: (
                r.usage.prompt_tokens,
                r.usage.completion_tokens,
            ),
        )

        result = await run.tool_call(
            fn=lambda: search_database(query),
            tool_name="search_database",
        )

except PikarcBlockedError as e:
    print(f"Blocked: {e.reason}")
🛡

Automatic Enforcement

Budget, concurrency, and kill switch checks on every call

📊

Full Cost Tracking

Token usage and costs calculated automatically per step

🔍

Complete Trace

Every step recorded with decisions, tokens, and duration

Built for Production AI Applications

Whether you're building autonomous agents or AI assistants, maintain control without sacrificing speed

Autonomous AI Agents

Let agents act independently while maintaining control over spending and execution.

Kill switch for emergency stops
Daily budget limits per agent user
Track every model call and tool use

Customer-Facing AI

Deploy AI assistants that interact with customers while keeping costs under control.

Per-user daily spending limits
Cost tracking per conversation
Block agents that exceed budgets

Multi-Agent Systems

Run multiple agents in parallel with resource controls and full visibility.

Concurrent run limits
Per-agent cost analytics
Full run timeline for debugging

Team AI Workflows

Give your team access to AI tools while maintaining budget governance.

Per-user budget enforcement
Usage analytics by user and model
Workspace-wide daily budget caps

Simple, Transparent Pricing

Choose the plan that fits your needs. All plans include core governance and observability features.

Basic

Free

Solo devs / Prototypes

Key Feature

Perfect for getting started

  • 1,000 runs/month
  • 3 concurrent runs
  • 1-day retention
  • Run tracking & cost analytics
  • Email support
Get Started

Pro

$25+ Usage

Growing AI Startups

Key Feature

Per-user daily budgets

  • 50,000 runs/month
  • Configurable concurrent limits
  • 14-day retention
  • Per-user daily budgets
  • Run tracking & cost analytics
  • Priority email support
Get Started

Scale

$79+ Usage

Production-ready Agents

Key Feature

Kill switch

  • Unlimited runs
  • Configurable concurrent limits
  • 90-day retention
  • Per-user daily budgets
  • Kill switch
  • Run tracking & cost analytics
  • Priority email support
Get Started

Enterprise

Custom

Large Org Governance

Key Feature

SSO, audit logs & dedicated support

  • Everything in Scale
  • Custom retention & run limits
  • SSO / SAML
  • Audit log export
  • Dedicated support & SLA
  • Custom deployment options
Contact Sales

All plans include core governance and observability features.

Start Governing Your AI Agents Today

Deploy AI agents with confidence. Budget enforcement, cost tracking, and full observability from day one.

Free tier available
No credit card required
Python SDK