Release 0.27

Release 0.27

This page records the notes of the 0.27 release.

Released 2026-10-07.

0.27.0 is a breaking release. The core plugin is built again from an empty tree. It uses only the documented extension points of Claude Code 2.1.292. These are a settings profile, a forced output style, one hooks module, agents, and skills. The Design page gives the source of each decision.

Migration from 0.26

• Install Node.js 22.18 or later. The scripts and hooks no longer run on Bun.
• Update, restart Claude Code, and run /dotclaude:setup. Setup removes includeGitInstructions: false, autoCompactWindow, and advisorModel. It also removes the 0.26 subagentStatusLine and its stub. It removes the 0.26 block in <config dir>/CLAUDE.md. These settings work against the new profile. Setup makes a backup of each file before it writes.
• The plugin options are gone. So the /config entries of dotclaude have no effect.
• Handoff notes from 0.26 in .claude/handoffs/ stay readable. No hook writes new ones.

Added

• Compaction is off: autoCompactEnabled: false and DISABLE_COMPACT=1. DISABLE_COMPACT=1 also turns off /compact.
• The context window is 300K tokens by CLAUDE_CODE_MAX_CONTEXT_TOKENS=300000. Claude Code honors that variable only together with DISABLE_COMPACT. At the limit the session stops, and the user runs /clear. CLAUDE_CODE_AUTO_COMPACT_WINDOW=300000 makes /context and the context warnings use the same window. Without it, Claude Code 2.1.292 uses 200K for Opus 5.5 there.
• Auto memory is off: autoMemoryEnabled: false.
• The forced dotclaude output style, with keep-coding-instructions: false. It works together with CLAUDE_CODE_SIMPLE_SYSTEM_PROMPT=1. The style replaces the built-in coding instructions and the 0.26 spec injection. The prompt cache holds it.
• The sandbox is on, with network access only to GitHub and the package registries.
• templates/managed-settings.json for users who want the compaction, memory, and sandbox keys enforced. Setup prints the copy command and does not run it, because it needs sudo. With it, the built-in guard sec-default loads. It skips the context_management text of the module.
• OpenSpec is a first-class integration. Setup checks the openspec CLI. It offers the install and openspec init --tools claude. The status line shows the active change with its task count.
• The profile turns off the tools that it does not use. So their schemas leave each request. The tools are enableArtifact: false, enableWorkflows: false, and deny rules for ScheduleWakeup and ReportFindings. /code-review then gives its findings as text. In the sandbox, /context at the start of a session fell from 32k to 8.1k tokens. The modder plugin was trimmed in the same run.
• The hook lab: just lab runs claude plugin test plugins/dotclaude. It loads the hooks module and fires stubbed tool.check events.

Changed

• Permissions ask only before dangerous commands. The rules are native, and there is no Bash parser. The dangerous commands are force pushes, history rewrites, piped shells, sudo, publishes, public gh writes, and SQL drops. Read-only git and the usual build and test commands are allowed.
• One hooks module replaces the 0.26 command hooks. It asks before a recursive rm outside the project. It also asks before the first call to a GitHub repository of another owner with an AI policy. It keeps a deny of the engine and fails open. With DISABLE_COMPACT set, its prompt.section handler replaces the context_management section of the system prompt. That section says that the system summarizes prior messages. The handler replaces it with text that says compaction is off. The built-in guard sec-default skips this handler on a machine with managed settings. It also skips it on a Team or Enterprise login.
• In auto mode, the classifier can allow an ask of the module. So one classic PreToolUse hook, hooks/auto-mode-guard.mjs, asks again for the same calls. It runs only in auto mode. It runs only for commands that its if filters match. It needs Node.js.
• The scripts and hooks of the core and dotclaude-jev plugins run on Node.js 22.18 or later, not on Bun. Setup changes a bun status line command to node. The um command of dotclaude-modder stays on Bun, because it uses the YAML parser of Bun. The allowed-tools of the second-opinion skill allow only node with jev.mjs. They do not allow each bun command. The drive-web-browser skill and the Browser page give npm install -g agent-browser.
• A prompt audit against Opus 5.5 removed two numeric reply caps. It removed them from the output style and from the reviewer agent. It also removed the Reason: line of the old spec format from the Jev note. The modder skill says that a /clear also removes what is not in MODLOG.md.
• /dotclaude:handoff is user-invoked only. It writes a note that primes the next session. The note holds the goal, the constraints, the decisions, the rejected approaches, and the proof.
• maxEffortLevel is xhigh. The profile sets no effort level. So each model keeps its default.
• The status line is one file. It shows the context against the 300K window. It is yellow at 75% and red at 90%.
• The add-on plugins work without the core plugin. dotclaude-browser is skill-only. dotclaude-jev gives its own session note.
• dotclaude-modder has one skill, mod-any-game, and no session note. The 8 other skills are step files in references/. Claude reads them only at their step. So the plugin adds one skill listing to each session, not nine. The rules of the session note are in the skill.
• includeGitInstructions stays at its default. The profile sets attribution.commit and attribution.pr.

Removed

• The operating spec, its SessionStart and SubagentStart injection, and lib/terms.mjs.
• The compaction fork, the cold-cache note, and each hook that tells the user to run /clear.
• The Bash, edit, secret-redaction, spawn, and attribution guards. The dependency on betterleaks goes too.
• The CodeGraph augment, the sembr rewrap, the long-run note, the setup notice, and the Stop prompt hook.
• The Ponytail integration and the minimal-code rule. The output style has its own rule to write the least code that does the task.
• All plugin options, the Concise output style, and the contribute skill.
• The agents translator, docs-writer, digest-writer, eval-designer, fuzz-engineer, infra-engineer, and reverse-engineer.
• The evals in plugins/dotclaude/evals/ and tools/compaction-report.mjs.
• advisorModel and CLAUDE_CODE_WORKFLOW_MAX_CONCURRENT_AGENTS. The advisor tool adds its instructions to each request. Each advisor call reads the whole context again.
• The reverse-engineer agent of dotclaude-modder. The reverse-engineer-binary skill in xsyetopz/skills replaces it.

Related Pages

• Previous: Release 0.26
• Next: Release 0.28
• Archive: the index of the archived pages.