Long jobs, held together
Claude models suit the multi-step reading, writing, and reasoning jobs agents spend most of their time on.
Anthropic
The lane your agents think on when the job calls for Claude — your key, Anthropic's prices.
model lane · your key, Anthropic's prices
Missions
What agents get done with it.
Your key does the paying
Anthropic bills the key you connect directly; Ahel adds nothing on top of it.
The picker follows the key
Once connected, the model picker fills with the Claude models your key can reach; you pick per agent.
The catalog
Every action, honestly listed.
A model provider gives an agent a lane to think on, not actions to call — so there is no action catalogue here. Tool actions come from the other integrations.
Connect
Your key, their prices.
- Your key
- pasted once in the desktop app · stored by your runtime, never by us · removable any time
- Markup
- none — Anthropic usage at the vendor’s own prices
- Auth
- your own Anthropic key, from Anthropic's console
- Billing
- Anthropic bills you directly — no markup
- Models
- the picker lists what your connected key reaches
- Scope
- thinking only — no tool actions ride this lane
Your own Anthropic key, created in their dashboard and pasted once into the Ahel desktop app.
FAQ
Questions people actually ask.
Where do I get an Anthropic API key?
From Anthropic's own developer console; paste it once in the Ahel desktop app and it stays in your runtime. Ahel is invite-only right now, with access opened in batches.
What does Claude cost through Ahel?
Exactly what Anthropic charges your key — Ahel adds no markup and sells no usage tier of its own. Anthropic's usage prices are theirs, on your Anthropic invoice.
Which Claude model should an agent use?
Whichever the job calls for; the picker shows the models your key reaches and you choose per agent. We don't rank them — Anthropic's own model docs are the honest source.
OpenRouter
OpenAILearn:What is an AI agent?