Know which agents are getting better — and which are just getting expensive.

A verdict is a ranked performance brief — not a blended cost × eval metric. Morse keeps each evidence lane visible so you can judge what moved and what still needs review. Why Morse ↓

1. 1

   Instrument

   Add the SDK — Python or TypeScript — or point an OpenTelemetry exporter at Morse. Typed agent, LLM, tool, HTTP, and database spans start flowing; telemetry batches asynchronously and never blocks the request.

   AnthropicVercel AI SDKOpenAI AgentsLangGraphLangChainor plain OTLP
   PythonTypeScript

   pip install morse-ai

   import morse_ai
   
   # One line — reads MORSE_API_KEY from your environment
   morse_ai.init()
   
   # Cost, quality, and traces appear as your agent runs

   Invites arrive by email during private beta; your API key is under Settings → API Keys. No credit card, now or at launch.

2. 2

   Map success

   Tell Morse what a successful run is for each agent — a confirmed booking, an extraction that passed validation, a ticket routed to the right queue. Outcomes can be labeled from your code, confirmed in the UI, or both. This is the step the brief is built on.

3. 3

   Read your first brief

   Start with the ranked list. Open a trace when you want to check the evidence. Next week's brief tells you whether the change helped.

   Demo org · illustrative data

   1. data-extractor45.2% · -8.1 ptsNeeds attention
   2. code-reviewer72.4% · +0.0 ptsWorth watching
   3. triage-agent79.4% · -11.9 ptsWorth watching