Start with one agent.
Scale with real usage.

Every plan includes the same core authorization, approvals, incidents, and audit capabilities. Choose automatic monthly or yearly billing; plan limits remain monthly.

No LLM token chargesNo automatic overagesFree developer entry
01

Basic

For one developer validating the integration.

$0USD
per month
2,000security decisions / month
1secured agent
2scoped API keys
  • Runtime ALLOW, DENY, and APPROVAL
  • Policy and tool synchronization
  • Approvals and incident visibility
  • Tamper-evident audit records
  • All supported Python integrations
Start free →
02

Plus

For pilots and small production workflows.

$9USD
per month

Or $108 billed yearly. Both options renew automatically until cancelled.

50,000security decisions / month
3secured agents
5scoped API keys
  • Runtime ALLOW, DENY, and APPROVAL
  • Policy and tool synchronization
  • Approvals and incident visibility
  • Tamper-evident audit records
  • All supported Python integrations
Request Plus →
04

Pro

For high-volume agent products and services.

$79USD
per month

Or $948 billed yearly. Both options renew automatically until cancelled.

2,000,000security decisions / month
30secured agents
50scoped API keys
  • Runtime ALLOW, DENY, and APPROVAL
  • Policy and tool synchronization
  • Approvals and incident visibility
  • Tamper-evident audit records
  • All supported Python integrations
Request Pro →

What counts as a decision? One unique authenticated authorization operation. Idempotent retries are not counted twice. Invalid credentials, health checks, and rate-limit rejections do not consume the monthly decision quota.

Core protection is not
a premium add-on.

Start free with the same enforcement semantics used by paid workspaces. Plans change capacity—not whether a denied operation stays denied.

01

Default deny

Unknown agents and undeclared tools do not inherit access.

02

Exact approvals

Human review binds to one action and its canonical arguments.

03

Framework adapters

Python, LangChain, LangGraph, CrewAI, and OpenAI Agents.

04

Operational evidence

Decisions, incidents, approval history, and audit integrity.

Choose for the workload
you operate today.

Begin on Basic and move up when additional agents, API keys, or monthly decisions are required.

PlanAgentsAPI keysDecisionsCore security
Basic122,000Included
Plus3550,000Included
Premium1020500,000Included
Pro30502,000,000Included
BUYING QUESTIONS

Clear limits.
No surprise behavior.

01Does a denied operation count?

Yes. ALLOW, DENY, and REQUIRE_APPROVAL are valid security decisions because the authorization service evaluated a real operation. Invalid authentication and health checks do not count.

02Does Quasentra charge for model tokens?

No. The deterministic authorization path does not invoke an LLM. Charges from your chosen model provider remain separate.

03What happens at the monthly limit?

New decisions fail closed with a quota response and reset date. Quasentra does not create an unexpected overage bill or silently allow the operation.

04Are integrations restricted by plan?

No. Supported Python, LangChain, LangGraph, CrewAI, and OpenAI Agents adapters use the same enforcement features on every plan.

05Can I choose monthly or yearly billing?

Yes. Monthly subscriptions renew every month. Yearly subscriptions charge the displayed annual total and renew each year. Both can be cancelled, and access changes only from verified payment events.

START WITHOUT A CREDIT CARD

Secure the first 2,000
decisions for free.

Connect one agent, verify the enforcement path, and inspect real authorization evidence before choosing a paid plan.

Create free workspace →