Pricing

Transparent, predictable pricing. Pay for what you use.

Plan

Feature Included
Monthly price $15/month
Agent profiles $2/month each
Monthly allowance $15 usage credit
Additional allowance Available on demand

Your $15/month subscription includes $15 of usage credit. Each agent profile is $2/month. The allowance is shared across all profiles.

Superpower Pricing

Category Superpowers Cost
External superpowers handler_research, handler_intel, handler_monitor, handler_finance, handler_communicate, handler_audio, handler_marketing $0.005/call
AI generation handler_generate 2x provider cost (e.g., GPT-4 cost x 2)
Connected service calls (BYOK) All OAuth-connected services (Gmail, GitHub, Slack, etc.) $0.0005/call
Real dollars, not credits. Handler pricing is in actual dollars, not abstract credits. When you see $0.005, that's half a cent. No conversion math needed.

What Counts as a Call

Each tool invocation is one call. A single call to handler_research with query: "AI agents" is one call, regardless of how many results come back.

Budget & Caps

Your $15 monthly allowance translates to approximately 3,000 external superpower calls at $0.005/call. Here's how budgets and caps interact:

Tip: Set a monthly cap per profile that's lower than your total allowance. This prevents one runaway agent from consuming your entire budget.

Example Usage

Here's what a typical month looks like for an active Handler user with one agent:

Activity Volume Cost
Research calls (search, read, answer) 500 calls $2.50
Intel calls (person search, company enrich) 200 calls $1.00
Monitor calls (news, twitter, reddit) 100 calls $0.50
Finance calls (quotes, indicators) 50 calls $0.25
Email sends 30 calls $0.15
AI generation 20 calls ~$0.40 (varies by model)
GitHub/Notion calls (connected services) 1,000 calls $0.50
Total 1,900 calls ~$5.30

At ~$5.30, this active usage is well within the $15 monthly allowance, leaving room for growth or spikes.

FAQ

What happens when I hit my cap?

Calls are blocked with a blocked status. Your agent receives a clear message explaining that the spending cap has been reached. No charges are incurred. Calls resume when the next billing cycle starts or when you increase the cap.

Is there a limit on agent profiles?

There is no fixed limit on agent profiles. Each profile is $2/month and gets its own key, governance rules, and budget controls.

Is there a free tier?

Yes — agents can self-register for a trial key with no credit card required. POST https://api.usehandler.dev/v2/keys/trial with any email and you get a sk-hndlr-* key immediately: $0.50 credit, 100 calls, 7-day TTL. See For AI Agents for details.

Do connected service costs count against my allowance?

Yes. Connected service calls ($0.0005/call) are deducted from your monthly allowance, just like external superpower calls. All usage is tracked in one place.

How is handler_generate priced?

handler_generate charges 2× the underlying provider cost. The default model is gpt-4o-mini (~$0.0003/1k input tokens at provider cost, so ~$0.0006 via Handler). Longer prompts cost more. Pass model in your call to use a different model — costs scale with the model's provider rate. The response always includes a cost field showing exactly what was charged.

Can I see my usage in real time?

Yes. Every Handler response includes a budget object showing your remaining balance and approximate calls remaining. You can also call handler_status at any time (free) to check your current usage and budget.

Enterprise

Custom pricing for teams that need SSO/IdP integration, dedicated support, SLA guarantees, and volume rates. Contact us.