Agents

Hosted agents, any uncensored model.

An agent is a model with a system prompt, webhook skills you host, memory per session and a schedule. Create it with one API call, message it with another, or from the console.

Create an agent

One turn, worked

POST /messages
session mod-1842
memory loaded for the session
Model call 1 · GLM-5.3 Uncensored
1,240 in → 46 out
$0.0004
Skill lookup_thread
signed POST → 200
Model call 2
1,410 in → 180 out
$0.0005
Turn done
tokens $0.0009 + fee $0.005
turn.done posted to your webhook, signed

Pricing

Starter
$9.00 a month
up to 3 skills, 1 cron
Pro
$39.00 a month
up to 20 skills, 20 crons
Per turn
tokens at the model price + $0.005
GLM-5.3 Uncensored per M
$0.282 in / $0.846 out

Fees from GET /v1/agents/pricing; tokens at the price on each model page. Both draw from the prepaid balance.

What an agent is

One model
Any chat model on the catalog, named by id. Change it with a PATCH; the memory and skills stay.
Your webhook skills
A name, a JSON schema and a URL you host. Tool calls arrive signed; your JSON answer goes back to the model.
Memory
Conversation history per session id, kept on the instance. Send the same session id to continue.
Cron
A prompt on a schedule, run as a turn. Skipped when the balance is empty.
API and webhooks
Send messages and read turns over REST. Turn events post to your webhook, signed, with retries.
Billing you can read
Every turn carries its charge, and each model call inside it is on the receipt.

Create one, then message it

create
1curl https://api.uncensoredllm.ai/v1/agents \
2 -H "Authorization: Bearer sk-ullm-live-…" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "name": "moderator",
6 "model": "glm-5.3-uncensored",
7 "tier": "starter",
8 "system_prompt": "You triage reports from our forum and answer in one paragraph.",
9 "skills": [{
10 "name": "lookup_thread",
11 "description": "Fetch a thread by its id",
12 "url": "https://example.com/hooks/lookup-thread",
13 "parameters": { "type": "object", "properties": { "thread": { "type": "string" } } }
14 }],
15 "webhook_url": "https://example.com/hooks/agent-events"
16 }'
message
1curl https://api.uncensoredllm.ai/v1/agents/AGENT_ID/messages \
2 -H "Authorization: Bearer sk-ullm-live-…" \
3 -H "Content-Type: application/json" \
4 -d '{ "session_id": "mod-1842", "content": "Is thread 10422 spam?" }'
  • The create call answers 201 with the agent and a webhook_secret, shown once; use it to check the signature on events and tool calls.
  • The message call answers with the turn: output, charge_usd, fee_usd, and the tool_calls and model_calls it took.
  • PATCH /v1/agents/:id with { "status": "stopped" } pauses it; POST /v1/agents/:id/crons with an expression and a prompt schedules it.

Questions

An agent is a model with a system prompt, a set of tools, and a memory, kept running for you. You send it a message; it calls your tools as many times as it needs, keeps the conversation per session id, and answers. It runs on the schedule you set and posts events to your webhook.

Any chat model on this catalog, by its id. The agent sends every model call through the same key and routing as a direct request, so the per-token price is the one on the model page.

A skill is a name, a description, a JSON schema for its arguments and a URL you host. When the model decides to use it, the agent POSTs the arguments to your URL with an HMAC signature and feeds your response back to the model. No OAuth, no SDK: any endpoint that answers JSON is a skill.

Two lines on the ledger. A flat monthly fee for the instance, $9.00 on Starter and $39.00 on Pro, and per turn the tokens at the model's normal price plus a $0.005 fee. Both draw from the prepaid balance; an unpaid monthly fee stops the agent until the balance covers it.

Limits. Starter runs with up to 3 skills and 1 cron; Pro with up to 20 skills and 20 crons. The model, the memory and the API are the same on both.

Yes. The loop (model call, tool call, model call again), the conversation memory, the scheduler, the retrying webhooks and the billing are the hosted part. Your side is the system prompt and the HTTP endpoints the skills call.

Chain models without an agent? See Workflows. Agents are created and watched in the console.