mirror of https://github.com/openai/codex.git synced 2026-11-14 21:40:39 +00:00

Files

T

Zahan Malkani 5943d7acdf Persist timestamped realtime frames before acknowledging completion (#61473 )

## What changed

- Record voice transcripts as `frame` items with a `voice_transcript` channel for eligible threads, while retaining legacy transcript support and the existing app-server live and mixed timeline representation.
- Assign first-observed timestamps that never move backward, retain them through completion, and restore the clock floor from history and compaction checkpoints.
- Flush realtime records before completion notifications, including file and initial directory durability barriers for local rollouts.
- Generate a distinct default session ID for each new call, use the call ID when attaching to an existing call, and derive opening record IDs from session IDs.
- Add protocol types for turn-input acceptance and session-end input receipts, excluding those receipts from live notifications and mixed timelines. Update generated schemas and SDK types.

## Testing

Add coverage for frame serialization and required voice payloads, clock rollback and checkpoint restoration, durability barrier failures and retries, per-call session IDs, and persisted transcript availability immediately after completion.

GitOrigin-RevId: 3f6632b517d563861aee085b723475b8945c391c

2026-11-14 06:14:36 +00:00

docs

Normalize formatting in docs, web assets, and sample scripts (#52256 )

2026-10-08 20:11:43 +00:00

examples

Embed friendly instructions in bundled GPT-5.4 and GPT-5.5 (#44930 )

2026-09-11 22:29:54 +00:00

notebooks

Embed friendly instructions in bundled GPT-5.4 and GPT-5.5 (#44930 )

2026-09-11 22:29:54 +00:00

scripts

Version answer display instructions and restrict unsupported payloads (#60904 )

2026-11-11 23:35:05 +00:00

src/openai_codex

Persist timestamped realtime frames before acknowledging completion (#61473 )

2026-11-14 06:14:36 +00:00

tests

Version answer display instructions and restrict unsupported payloads (#60904 )

2026-11-11 23:35:05 +00:00

_runtime_setup.py

Support alpha hotfix release versions (#34463 )

2026-07-21 03:59:38 +00:00

BUILD.bazel

Add Bazel targets for on-demand schemas and Python SDK generation (#54096 )

2026-10-14 20:59:31 +00:00

pyproject.toml

Bump the Python SDK build backend to uv_build 0.13 (#59512 )

2026-11-04 23:05:12 +00:00

README.md

Add untrusted external messages to the Python SDK (#44086 )

2026-09-09 06:54:18 +00:00

release_version.py

Build Python SDK artifacts before publishing the runtime (#44061 )

2026-09-09 05:14:27 +00:00

RELEASING.md

Publish Python packages after stable CLI releases (#44067 )

2026-09-09 05:39:52 +00:00

uv.lock

Upgrade Python SDK model generation to datamodel-code-generator 0.64.0 (#56272 )

2026-10-22 21:40:58 +00:00

README.md

OpenAI Codex Python SDK

Build Python applications that start Codex threads, run turns, stream progress, and control workspace access.

Install

Install the SDK:

pip install openai-codex

Quickstart

The SDK reuses your existing Codex authentication when one is already available:

from openai_codex import Codex

with Codex() as codex:
    thread = codex.thread_start()
    result = thread.run("Explain this repository in three bullets.")
    print(result.final_response)

thread.run(...) returns a TurnResult containing the final response, collected items, and token usage.

Authentication

Existing Codex authentication is reused automatically. To start ChatGPT browser login explicitly:

from openai_codex import Codex

with Codex() as codex:
    login = codex.login_chatgpt()
    print(login.auth_url)
    print(login.wait().success)

For device-code login:

with Codex() as codex:
    login = codex.login_chatgpt_device_code()
    print(login.verification_url, login.user_code)
    login.wait()

For API-key login:

with Codex() as codex:
    codex.login_api_key("sk-...")

Built-In Help

Use Python's standard help(openai_codex), help(Codex), or python -m pydoc openai_codex documentation tools.

Documentation

The package is licensed under the repository Apache License 2.0.