Bring agentic possibilities to life.

A bonsai growing from a floating island of forest floor

Cadenya is the agent runtime that layers tools, agents, and objectives so you can test safely and improve quickly.

Sign up See API docs

Start with your stack

Connect your MCP servers, OpenAPI specs, and existing endpoints through a single tool layer agents can use.

Iterate without uprooting

Swap models, evolve behaviors, and expand capabilities while retaining infrastructure.

Evolve with what’s next

Adopt frontier models fast. Test behaviors and compare approaches. Cadenya’s unified runtime lets you add functionality, not complexity.

01.

Define an agent

02.

Shape abilities

03.

Run objectives

Start with one agent
grow on
and on and on.

Cadenya is the agent runtime that layers tools, agents, and objectives so you can test safely and improve quickly.

Default

Anthropic Claude Sonnet [Latest]

Canary

OpenAI: GPT-5.5

new variation

Canary

Model

GPT - 5.2

Created

2 days ago

Assignments

  • reroute shipment tool
  • update_eta tool
  • Dispatch API tool set
  • Customs Broker sub-agent

Memory layers

Carrier Playbook - SLA Policies

System prompt

You are Meridian’s shipment-exceptions agent. When a delivery stalls, reroute it via the dispatch API...

Stay on top of costs with live token metering. Reduce waste through progressive discovery, and improve efficiency as agents adapt.

Enable progressive tool discovery

Max tools per search

1-10

5

Search hints

up to 5

delays, reroutes, customs

Rerank threshold

0.0-10 [Leave blank to skip reranking]

Tool schemas stay out of the context window until the agent asks for them — only names ride along, so every request gets smaller.

Use webhooks and SSE to push live updates. Cadenya makes it easy to wire agent events into your apps.

webhook deliveries

Search url
Created Event type HTTP URL Status
Just now Assistant message 204 https://app.meridianfreight.com
/hooks/agents Completed
Just now tool_result 204 https://app.meridianfreight.com
/hooks/agents Completed
Just now tool_approval_requested 204 https://app.meridianfreight.com
/hooks/agents Completed
Just now sub_agent_spawned 204 https://app.meridianfreight.com
/hooks/agents Completed
Just now context_window_compacted 204 https://app.meridianfreight.com
/hooks/agents Completed
Just now timed_out 204 https://app.meridianfreight.com
/hooks/agents Completed

Monitor outcomes and understand how behaviors take shape in the real world. Clear visibility means your agents show their worth.

Feedback

Search comments All sentiment All variations
Created Comment Variation objective
+1.00 Rerouted before the SLA breach-recipient never noticed. Canary | GPT-5.2 view objective
+0.50 Caught the customs hold and updated the ETA proactively. Default | Claude Sonnet 5 view objective
+0.50 Good reroute, but notified the recipient twice. Canary | GPT-5.2 view objective
-0.25 Held at facility when a reroute was available Canary | GPT-5.2 view objective
+0.75 Escalated the frozen-goods lane correctly Default · Claude Sonnet 5 view objective

Just
add water

Kick off an agent using APIs you already have. It only takes a few steps—the first one is signing up.

Sign up
A lone tree on a rocky island, mirrored in still water

Frequently
asked questions

Question 01 What is Cadenya?

Cadenya is an agent runtime. It layers tools, agents, and objectives on top of the APIs you already run, so you can build, test, and improve agents without rebuilding your stack.

Question 02 Do I need to rewrite my APIs to use it?

No. Connect MCP servers, OpenAPI specs, and existing endpoints through a single tool layer, and agents can use them as they are.

Question 03 Which models can my agents run on?

The frontier model that fits each job. Variations let you run the same agent on different models side by side and compare behavior before you commit.

Question 04 How do agents stay within budget?

Live token metering shows spend as it happens, and progressive tool discovery keeps schemas out of the context window until an agent asks for them.

Question 05 How do I know what an agent actually did?

Every objective keeps its trail: tool calls, webhook deliveries, token usage, and the feedback people leave on the outcome.

Question 06 When can I start building?

Cadenya is not live yet. Sign up and we will let you know the moment it is.

A sprawling bonsai on a floating island trailing roots

Spin up
something wild.

Then watch your agents thrive.
Start building with Cadenya today.

Create an agent See API docs