product updateOpenAI

OpenCode v2.3.1 moves timed-out shell commands to the background and enables background compaction by default

TL;DR

OpenCode v2.3.1 now moves shell commands that hit their timeout to the background instead of killing them. Background compaction is on by default, and GPT 6.2 Sol and GPT 6.2 Astra appear in the model picker for ChatGPT sign-in.

3 min read
0

OpenCode v2.3.1 changes how the agent handles long-running shell commands: a command still running when its timeout expires now moves to the background instead of being killed, in every build. The release notes are published on the project's GitHub releases page (anomalyco/opencode).

Highlights

  • Timed-out shell commands keep running. Long builds and test runs can finish and report back. The shell_timeout_background flag is removed, though configs that still set it keep loading. A foreground timeout: 0 now uses the default 120 second timeout instead of waiting until the command exits (#63699).
  • Background compaction is on by default. As a session nears automatic compaction, the summary is written in the background so the next prompt does not wait for it. Set compaction.background: false to turn it off. A background summary that is never used still costs tokens. The experimental.background_compaction setting is removed (#64290).
  • GPT 6.2 with ChatGPT sign-in. GPT 6.2 Sol and GPT 6.2 Astra, plus their fast variants, are added to the ChatGPT model list (#64508).
  • Plugin updates from Settings. Open Settings > Extensions, click Check for updates, then Update on any plugin with a new version (#64511).
  • Comment on selected code. Selecting code in diff and file viewers now just selects it, so it can be copied, and an Add comment toolbar appears beside the selection (#64410).

Other core changes

  • OpenCode again reads CLAUDE.md in any directory that has no AGENTS.md, and falls back to ~/.claude/CLAUDE.md when the global AGENTS.md is missing. This also applies to instructions found when the agent reads nested files. Set OPENCODE_DISABLE_CLAUDE_CODE or OPENCODE_DISABLE_CLAUDE_CODE_PROMPT to skip every CLAUDE.md (#64474, #64533).
  • Experimental schedules are now automations, behind the automations flag (the old schedules flag still works). An automation runs on a time, an interval, a cron or a pushed MCP event. It either sends a prompt (to the creating session, another session or a new subagent) or runs a Code Mode program that decides whether to send one. Use /automation or the automation_* tools. Active or paused plain prompt schedules carry over; schedules with a condition or run limit, and finished ones, are dropped (#64160).
  • websearch now returns a flat list of results, each with url, title, content and published date (#64471).
  • read gives Code Mode and plugins a simpler structured result: text (with offset, lineCount and next), directory (entry names) or media (a URI instead of base64). What the model sees is unchanged (#64361).
  • Code Mode programs can use Node's Buffer, search("issues") works as shorthand for search({ query: "issues" }), and tool summaries in the catalog can be up to 500 characters instead of 120 (#64365, #64347, #64341).
  • glob and grep now report a malformed filename glob (for example [) as an error naming the parameter to fix, instead of returning no results (#63960).
  • Server requests that crash return a 500 with a readable JSON UnknownError body. Storage problems name the cause, such as "Storage is full" or "Storage is read-only". Other crashes include a ref that can be found in the server log (#64022).
  • The shell tool description now tells the model that its session ID is in the OPENCODE_SESSION_ID environment variable (#63718).

Bug fixes

Finished shell commands no longer leave zombie processes under opencode-pty. The process daemon now cleans each command up when it exits.

What this means

The timeout change affects any workflow where the agent runs builds or test suites that outlast the timeout: those commands now continue and report back rather than being terminated. The compaction change removes a wait before the next prompt, with a stated trade-off: a background summary that goes unused still consumes tokens, and users who want to avoid that cost can disable it with compaction.background: false. The CLAUDE.md fallback matters for projects that carry instructions in that file without an AGENTS.md. Teams using the experimental schedules feature should check their existing schedules before upgrading, since those with a condition or run limit are dropped in the move to automations.

Related Articles

product update

Perplexity Brain Can Now Create and Update Agent Skills; 80,000+ Personal Skills Written for 20,000+ Users

Perplexity said on November 18, 2026, that Brain, the personalization system for its Computer agent, can now create and update reusable agent skills. According to the company, Brain wrote more than 80,000 personal skills for over 20,000 users between August 1, 2026, and November 14, 2026.

product update

GitHub Copilot team customizations now generally available for organization teams

GitHub announced that team customizations for GitHub Copilot are generally available for organization teams. The feature lets teams share the Copilot plugins they rely on without asking every member to repeat the same setup.

product update

Google adds "temporary threads" to AI Mode in Search, detaching chats from your account

Google is adding "temporary threads" to AI Mode in Search, a chat option that detaches the session from the user's account. The conversations won't resurface in search history or recommendations. The feature is rolling out globally on desktop and mobile across all supported languages.

research

OpenAI Reports Up to 14% PPA Gain on AMD Flagship GPU IP From Model-Rewritten RTL

OpenAI reported on November 18, 2026 that its models achieved up to a 14% improvement in power, performance and area (PPA) on AMD flagship GPU intellectual property, measured against heavily optimized baselines. Some of the circuit rewrites were included in a design sent for manufacturing.

Comments

Loading...