# Google open-sources the Elephant-Goldfish Model plugin its engineers use to catch AI coding mistakes

DevFeed: [Google open-sources the Elephant-Goldfish Model plugin its engineers use to catch AI coding mistakes](<https://devfeed.tech/articles/google-open-sources-the-elephant-goldfish-model-plugin-its-engineers-use-to-catch-ai-coding-mistakes-99699.md>)

Original publisher: [Read original article](<https://www.helpnetsecurity.com/2026/10/26/google-elephant-goldfish-model-plugin/>)

Author: Anamarija Pogorelec

Published: 2026-10-26T04:25:59Z

Content type: news

Language: en

Sources: [Help Net Security](<https://devfeed.tech/sources/help-net-security.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [Documentation](<https://devfeed.tech/topics/documentation.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-code](<https://devfeed.tech/tags/ai-code.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-agent](<https://devfeed.tech/tags/ai-coding-agent.md>), [antigravity](<https://devfeed.tech/tags/antigravity.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [code](<https://devfeed.tech/tags/code.md>), [google](<https://devfeed.tech/tags/google.md>), [mistakes](<https://devfeed.tech/tags/mistakes.md>), [news](<https://devfeed.tech/tags/news.md>), [open](<https://devfeed.tech/tags/open.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [plugin](<https://devfeed.tech/tags/plugin.md>)

## AI overview

Google's open-source Elephant-Goldfish Model plugin adds a design review gate before coding: a persistent AI session interviews the developer and drafts a Markdown plan, while fresh, stateless sessions check whether they understand it, hunt for gaps, and make a final go/no-go call on whether implementation details are clear. If reviewers find ambiguity or objections, the plan returns for revision. The plugin includes customizable reviewer personas and workflows, but security flaws are not explicitly part of the default review; teams must add a security-focused reviewer themselves. It currently works in Google's Antigravity environment, with support for others described as forthcoming. Google says thousands of employees use it daily for production code.

## Source excerpt

Google has open-sourced the Elephant-Goldfish Model plugin, an internal tool that makes an AI coding agent write and defend a design document before it touches any code. The plugin hands that document to fresh AI sessions that have never seen the design conversation and tells them to find what is wrong with it. The EGM plugin in a nutshell (Source: Google) "AI-powered velocity is transformative; however, it has silently introduced a potential reliability crisis by ... More -> The post Google open-sources the Elephant-Goldfish Model plugin its engineers use to catch AI coding mistakes appeared first on Help Net Security.