September 2026 Summaries

Entire CLI now integrates with Google’s Antigravity coding agent, automatically tracking prompts, session transcripts, file changes, token usage, and associated commits while users continue their usual agent workflow. The integration enables teams to review the conversations behind AI-generated code through session reviews and Agentic Search, and joins support for agents including Claude Code, Gemini CLI, Cursor, Copilot CLI, Codex, and others. Entire emphasizes decentralized code storage by region and continued agent use during repository outages elsewhere. Users can enable Antigravity in existing or new repositories through CLI commands and verify tracking or configuration with status and diagnostic commands. Additional updates make large-repository mirrors initialize asynchronously, add a PowerShell installer for Windows, respect Git’s native reference locks to prevent conflicting checkpoint updates, and improve checkpoint reliability in worktrees and nonstandard Git configurations.

Entire CLI has added integration with Google’s Antigravity coding agent, enabling automatic capture of agent sessions, prompts, transcripts, file changes, token usage, and links to related Git commits. Developers can continue using Antigravity normally while Entire records background context that can be used for branch reviews by multiple agents or searched through Agentic Search to understand the conversations behind code changes. The decentralized platform allows code to remain in a chosen region and supports work during outages affecting other repository locations, while Antigravity joins supported agents including Claude Code, Gemini CLI, Cursor, Copilot CLI, Codex, and others. Setup involves installing and authenticating Antigravity, enabling it in an Entire repository through CLI commands, and checking configuration or tracking status with diagnostic commands. Additional updates make large-repository mirroring asynchronous, add a PowerShell installer for Windows, improve protection against concurrent Git reference updates, and increase checkpoint reliability in worktrees and nonstandard Git repository configurations.

Entire’s Dispatch 0x001D announces Entire-native branches, which allow users to keep pushing code to Entire’s distributed Git network during GitHub outages by using the `entire/unmirrored/` branch prefix; these branches remain regional, are not synchronized back to GitHub, and can later be merged. The update also improves Git network performance and reliability by removing packfile read locking, enabling more frequent incremental repository maintenance for faster clones and fetches, releasing repository names left reserved after failed mirror creation, and allowing interrupted large repository clones in git-sync to resume automatically. Additionally, a go-git security fix prevents credentials from being sent to unintended redirect destinations, while Entire’s file viewer now displays a clear notice rather than unreadable characters for binary files such as PDFs and ZIP archives.

Entire’s Dispatch 0x001D introduces Entire-native branches, which let users continue pushing code during GitHub outages by storing ephemeral, region-local branches under the entire/unmirrored/ prefix without syncing them back to GitHub until they are merged later. The update also improves Git network performance and reliability by removing packfile read locking for faster concurrent cloning, enabling incremental repository maintenance for quicker clones and fetches, releasing repository names after failed mirror creation attempts, and allowing git-sync to resume interrupted large-repository clones. Additional changes strengthen go-git redirect request security and replace unreadable binary-file previews in Entire’s web viewer with a clear notice that binary files cannot be displayed.

Entire has introduced Agentic Search, an API and CLI-integrated tool that lets coding agents search across accessible repositories for both exact code and the historical reasoning, commits, and session transcripts behind changes. Its dual search engine combines semantic retrieval of discussions and commit context with symbol-aware code search, using turbopuffer for hybrid BM25 and semantic queries with regionally distributed storage, while code indexes are incrementally maintained and cached for roughly 100-millisecond median query latency. In internal tests involving nine engineering-history questions and 90 runs per scenario, agents using search answered 81 of 90 questions correctly versus 70 without it, while reducing average token use from 547,000 to 262,000, costs from $0.38 to $0.23, and average steps from 14 to 7; however, the company notes that the benchmark may favor its own detailed, agent-written commit history. Users can enable the feature through the Entire CLI, search sessions, commits, transcripts, and code across selected repositories, use browser or JSON interfaces, and access definitions and references, as Entire expands its platform as a record of agentic coding work.

Entire has introduced Agentic Search, an API and CLI feature that lets coding agents search across accessible repositories for both exact code and historical context from commits, sessions, and transcripts. The system combines semantic search for understanding why changes were made with symbol-aware code search for locating definitions, references, and specific lines, while supporting searches across individual or organizational repositories. Its indexing pipelines process repository events, store semantic data with turbopuffer for combined BM25 and nearest-neighbor retrieval, and maintain code indexes in S3 and local caches, reporting median query latency of about 100 milliseconds. In internal benchmarks involving nine engineering-history questions, agents using search answered 81 of 90 runs correctly compared with 70 without it, while using fewer tokens, steps, and less time, though the company notes that its own detailed commit history may bias the results. Users can enable the feature through the Entire CLI, use command-line or browser interfaces, and obtain structured results for agent workflows.