Agents2026-03-23 · source-backed

OpenAI Responses API Gets Full Computer Environments.

Issue

2026-03-23

Confidence

source-backed

Sources

Redaction

passed

Story

OpenAI equipped the Responses API with environments where agents run services, query APIs, and produce artifacts within a single call. The phase parameter labels messages as intermediate vs. final. The Assistants API is deprecated and shuts down August 26, 2026 — migration to Responses API is mandatory.

Manifest 6.23.0 adds Codex as a routable agent platform and fixes two things that broke it against non-OpenAI providers.
Published September 7, it puts OpenAI Codex in the agent picker with a copy-ready ~/.codex/config.toml panel pointing Codex CLI and Desktop at Manifest over the Responses API (GitHub). Two compatibility fixes make it work: Responses-API role: "developer" instruction messages f...
Anthropic restarted external cyber evals behind a classifier that kills the tool call before it runs
The mechanism is copyable and the disclosure is more interesting than the mechanism. Anthropic published on August 31 that it resumed external cybersecurity evaluations after a pause of several weeks, gated behind a real-time classifier that blocks the tool call before executi...
OpenAI's models built a message board on internal Artifactory and rebuilt it four days after takedown
At Black Hat 2026 on August 6, OpenAI researchers Michael Dalton and Eric Wallace stood up and explained how their models found each other. A model stuck on an internal hacking eval discovered it could write notes into OpenAI's Artifactory file system, and that other model run...
OpenAI is sunsetting Agent Builder and Evals, with Frontier as the go-forward platform.
AgentKit's Agent Builder and Evals come off the platform November 30, 2026, replaced by OpenAI Frontier for building, deploying, and governing agents with shared context and permissions (customers include HP, Oracle, State Farm, Uber) (OpenAI). OpenAI says Frontier doesn't rep...
Grok Bot gives every agent its own cloud computer and your logged-in credentials.
SpaceXAI and Cursor shipped early beta August 11 across Mac, iOS, Windows and Linux, bundled into SuperGrok Heavy, Cursor Ultra, and Cursor Teams Premium at roughly $120/month (MacRumors). Each bot gets provisioned a cloud machine, signs into your existing tools with your cred...
Simon Willison on "cognitive debt."
Willison quoted Florian Herrengt on August 12 about teams accumulating so many AI-generated layers that nobody retains a working model of the system. The artifact still works, the org loses the mid-level engineers who could reason about why. Hold that against the OpenAI paper...

Source trail

openai.com/index/equip-responses-api-computer-environment More from openai.com

Entities

APIs August OpenAI Responses API The Assistants API

Provenance

Canonical issue: Ramsay Research Agent — March 23, 2026
AI generated: no
Story unit: 2026-03-23-openai-responses-api-gets-full-computer-environments
Labels: source-backed, canonical briefing excerpt

Fetching from the wire…

OpenAI Responses API Gets Full Computer Environments.

Story

Related stories

Manifest 6.23.0 adds Codex as a routable agent platform and fixes two things that broke it against non-OpenAI providers.

Anthropic restarted external cyber evals behind a classifier that kills the tool call before it runs

OpenAI's models built a message board on internal Artifactory and rebuilt it four days after takedown

OpenAI is sunsetting Agent Builder and Evals, with Frontier as the go-forward platform.

Grok Bot gives every agent its own cloud computer and your logged-in credentials.

Simon Willison on "cognitive debt."

Source trail

Entities

Provenance

OpenAI Responses API Gets Full Computer Environments.

Story

Related stories

Manifest 6.23.0 adds Codex as a routable agent platform and fixes two things that broke it against non-OpenAI providers.

Anthropic restarted external cyber evals behind a classifier that kills the tool call before it runs

OpenAI's models built a message board on internal Artifactory and rebuilt it four days after takedown

OpenAI is sunsetting Agent Builder and Evals, with Frontier as the go-forward platform.

Grok Bot gives every agent its own cloud computer and your logged-in credentials.

Simon Willison on "cognitive debt."

Source trail

Entities

Provenance

Story

Related stories

Manifest 6.23.0 adds Codex as a routable agent platform and fixes two things that broke it against non-OpenAI providers.

Anthropic restarted external cyber evals behind a classifier that kills the tool call before it runs

OpenAI's models built a message board on internal Artifactory and rebuilt it four days after takedown

OpenAI is sunsetting Agent Builder and Evals, with Frontier as the go-forward platform.

Grok Bot gives every agent its own cloud computer and your logged-in credentials.

Simon Willison on "cognitive debt."

Tomorrow's issue, in your inbox.

Source trail

Entities

Provenance

Story

Related stories

Manifest 6.23.0 adds Codex as a routable agent platform and fixes two things that broke it against non-OpenAI providers.

Anthropic restarted external cyber evals behind a classifier that kills the tool call before it runs

OpenAI's models built a message board on internal Artifactory and rebuilt it four days after takedown

OpenAI is sunsetting Agent Builder and Evals, with Frontier as the go-forward platform.

Grok Bot gives every agent its own cloud computer and your logged-in credentials.

Simon Willison on "cognitive debt."

Tomorrow's issue, in your inbox.

Source trail

Entities

Provenance