DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to stress-test a design doc before writing code with Elephant Goldfish Model

Run a multi-stage Goldfish reviewer swarm over a draft spec to find gaps before implementation.

Who it is forA software engineer about to start a feature who has a written design doc and wants independent review before coding begins.

The problemDesign docs often have gaps that only show up during implementation, and the author is too close to the text to see them.

The outcomeA reviewed design doc with findings from several isolated reviewer passes, ready to be resolved and submitted by pull request.

What you need first

  • An AI coding agent that can install the plugin, such as Claude Code or Codex, or any agent that can take the install request from the README
  • A local git checkout of the project you are designing for
  • A design doc file under docs/designs/ to review, or a concept to start from

Steps

  1. Install the plugin in Claude Code with claude plugin marketplace add google/elephant-goldfish-model && claude plugin install egm@egm
  2. Start Claude Code and run /egm:egm-review docs/designs/<design-doc>.md on your draft (inferred, check the README)
  3. Read the findings from each stage: Zoom Out, Comprehension, Critic, and Readiness
  4. Run /egm-resolve to triage the findings and apply patches you approve
  5. Submit the verified design doc via pull request, then implement against it

Prompt for your agent

In the repo <repo-path>, use the EGM plugin's /egm-review skill on docs/designs/<design-doc>.md. Report the findings grouped by review stage and list which findings the design doc does not yet answer. Verify that each finding cites a section of the draft that exists.

Caveat. Goldfish reviewers read only the draft spec and live source code, so they cannot see decisions or context that live outside those files.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page