Is the OpenAI Agents SDK open source?

Yes. Both of OpenAI’s own packages, openai-agents-python and openai-agents-js, ship under the MIT license, so the SDK’s source is public, modifiable, and redistributable without asking OpenAI first. That covers the wrapper only: the loop, handoffs, guardrails, and tracing the SDK adds on top of a model call. What the loop calls, OpenAI’s models through the Responses API, is unaffected by the license on the code that calls it; running an agent still bills like any other OpenAI usage, open source or not. The Claude Agent SDK’s own license story is only half the same: its Python package is MIT-licensed too, but its TypeScript package is all rights reserved under Anthropic’s commercial terms, a per-language split this SDK doesn’t have. Either way, reading or changing the wrapper’s source doesn’t change what running it costs.

sources

keep reading

More on this.

Two ways to run Tessary.

Tessary is an open-source agent reliability platform. Cloud and self-hosted run the same workflow on the OpenTelemetry traces your agent already emits.

Tessary Cloud

We host it for you. Send your first trace with nothing to deploy and no model key.

what's includedper organization
traces
10,000 per calendar month
stored trace data
1 GB
retention
30 days
model credit
$10, one-time, for triage and root-cause analysis
credit card
not required

Self-hosted Tessary

Run the open-source code on your own infrastructure with one command. Add your own model key for triage and root-cause analysis.

Self-host Tessary for me by following https://github.com/tessaryai/tessary/blob/main/setup.md

docker compose -f oci://docker.io/tessaryai/tessary:compose up -d -y