Skip to content
AI News HubLIVE
More
Summary only

Codex Harness in OpenClaw

OpenClaw now uses the Codex app-server harness as the default runtime for OpenAI agent turns, separating the model loop from the product layer. This reduces tool duplication, prompt bloat, and translation overhead, bringing dynamic tool loading, deliberate visible replies, structured heartbeats, and subscription-based auth. Future improvements will benefit all model providers.

SourceHacker News AIAuthor: sjf
Codex Harness in OpenClaw
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions

Summary only

Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.

Key points and analysis

Article intelligence

EngineersBeginner

Key points

  • OpenClaw delegates the OpenAI model loop to Codex, while owning the product layer (channels, persona, memory, etc.).
  • Dynamic tool search and on-demand loading reduce initial prompt size significantly.
  • Replies and heartbeats are now intentional through dedicated tools, avoiding accidental text leakage.
  • ChatGPT subscription auth is supported, and Codex state is isolated per agent.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Agents