Skip to content

v0.3.0 — OpenAI SDK auto-instrumentation

Choose a tag to compare

View all tags
@arpitcoder arpitcoder released this 15 May 18:32
· 13 commits to main since this release
v0.3.0
4647cad

The adoption story. AEGRAIL_INTERCEPT=1 + Agent.from_env() and every openai.chat.completions.create / openai.responses.create (sync and async) is automatically wrapped with budget pre-check, audit emission, and token tally — no code change in the calling program.

Streams pass through. Cost stays caller-provided per CLAUDE.md design principle. Anthropic + litellm fast-follows next.

8 new integration tests, battle-tested against the built wheel in a fresh venv. Total suite: 161 tests, all green on 3.10 / 3.11 / 3.12.

Assets 2
Loading