Interfaces + Open Design / Foundation

Google's NEW EGM Skill Set + Deepseek is ACTUALLY INSANE!

This video explains Google's Elephant Goldfish Model workflow for designing non-trivial software features before coding: a context-rich elephant develops the plan, fresh goldfish agents expose hidden assumptions, and a human resolves findings and approves the tradeoffs. It covers the complete path from brainstorm and concept brief through iterative drafting, staged review, and resolution, plus the time and token costs that make EGM unsuitable for small, clear fixes.

AICodeKing12 minTranscript found

Quick learning frame

Read this before watching.

Coding-agent workflow is the loop of inspect, route, plan, edit, verify, summarize, and decide what should be automated next.

New playlist item from AICodeKing; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Skill you build: The ability to guide a non-trivial feature through a complete design lifecycle using a context-rich drafter, isolated skeptical reviewers, and explicit human approval before implementation.

Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.

Concept diagram

Where this video fits.

01Inspect context
02Route tool
03Plan work
04Edit safely
05Verify behavior
06Report next step

Deep lesson

Turn this video into working knowledge.

2,412 cleaned transcript words reviewed across 753 timed caption segments.

Thesis

Google's NEW EGM Skill Set + Deepseek is ACTUALLY INSANE! teaches a practical coding-agent workflow move: This video explains Google's Elephant Goldfish Model workflow for designing non-trivial software features before coding: a context-rich elephant develops the plan, fresh goldfish agents expose hidden assumptions, and a human resolves findings and approves the tradeoffs. It covers the complete path from brainstorm and concept brief through iterative drafting, staged review, and resolution, plus the time and token costs that make EGM unsuitable for small, clear fixes.

The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.

0:19

Split Agent Contexts

“twice is where it runs. It works in Antigravity, which is Google's own agent. But, it also has first-class install instructions for Claude Code, Codex, Claude Cowork, and even ChatGPT work. And because it's all just skills and...”

The elephant retains the requirements, conversation, and prior decisions while writing the design; each goldfish starts fresh with only that document and its referenced files. This isolation exposes missing definitions, failure behavior, and other gaps hidden by shared assumptions, while the human remains responsible for tradeoffs and final approval. Give one feature design to a fresh, read-only agent and list three assumptions it questions that the original design conversation left implicit.

3:17

Run The Lifecycle

“go. Now, let's talk about the workflow. In Claude code, all the commands use the /egm: prefix. The first one is /egm:egm brainstorm. You give it a feature, something like add webhook delivery to this project. The agent...”

EGM brainstorm sends six reviewers across lenses such as necessity, existing alternatives, and cost, then records the chosen direction in concepts brief.md. Draft writes the design section by section for human feedback; review applies five passes from architectural simplification through implementation readiness; resolve lets the human decide which findings to adopt before approving the design for coding. Take one proposed feature through a short brainstorm, concept brief, section-by-section draft, five-lens review, and written resolution decision.

8:53

Use It Selectively

“and Klein can do exactly the same thing. So, clone the whole repo, not just the skills folder, point Klein at the skill you want, and let it spawn fresh sub-agents for the goldfish. Just make sure the...”

EGM is meant for features with unresolved design decisions, not small, clear bug fixes where design and review overhead will not pay off. Its isolated reviewers repeatedly reread the design and code, so stronger assumption testing comes with substantial time and token costs. Classify two upcoming changes as either direct implementation or EGM candidates, citing the unresolved decision and expected review cost for each.

01

Inspect context

Start with this video's job: This video explains Google's Elephant Goldfish Model workflow for designing non-trivial software features before coding: a context-rich elephant develops the plan, fresh goldfish agents expose hidden assumptions, and a human resolves findings and approves the tradeoffs. It covers the complete path from brainstorm and concept brief through iterative drafting, staged review, and resolution, plus the time and token costs that make EGM unsuitable for small, clear fixes. Treat "Inspect context" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:19, where the video says: “twice is where it runs. It works in Antigravity, which is Google's own agent. But, it also has first-class install instructions for Claude Code, Codex, Claude Cowork, and even ChatGPT work. And because it's all just skills and...”

02

Route tool

Use "Route tool" to locate the part of the coding-agent workflow mechanism the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 3:17, where the video says: “go. Now, let's talk about the workflow. In Claude code, all the commands use the /egm: prefix. The first one is /egm:egm brainstorm. You give it a feature, something like add webhook delivery to this project. The agent...”

03

Plan work

Turn "Plan work" into the reusable artifact for this lesson: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal. This is where watching becomes something you can inspect and reuse.

04

Edit safely

Use "Edit safely" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.

05

Verify behavior

Use "Verify behavior" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.

06

Report next step

Use "Report next step" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.

Example

Source-backed artifact packet

Convert the video into a scoped artifact request that includes the transcript claim, mechanism, acceptance criteria, and proof. The output should be a coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal..

Example

Coding-agent workflow proof brief

Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the coding-agent workflow pattern.

Example

Teach-back module

Transform the lesson into a definition, a Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step diagram, one misconception, one practice exercise, and a check-for-understanding question.

Do not learn it wrong
  • Treating the title as the lesson without checking what the transcript actually says.
  • choosing tools by hype
  • losing context across agents
  • letting parallel sessions become invisible
  • Letting the lesson drift into generic Codex vs Claude comparison.
  • Letting the lesson drift into feature lists without task routing.
  • Letting the lesson drift into claims that ignore limits or recovery.

Transcript-derived moments

Use timestamps to study the actual video.

Quality check

Do not count this as learned until these are true.

01

State the transcript-backed claim in your own words: This video explains Google's Elephant Goldfish Model workflow for designing non-trivial software features before coding: a context-rich elephant develops the plan, fresh goldfish agents expose hidden assumptions, and a human resolves findings and approves the tradeoffs. It covers the complete path from brainstorm and concept brief through iterative drafting, staged review, and resolution, plus the time and token costs that make EGM unsuitable for small, clear fixes.

02

Explain the practical stakes without hype: New playlist item from AICodeKing; queued for transcript-backed review, topic mapping, and a practical learning artifact.

03

Map the idea onto the Intent -> Context -> Generation surface -> Preview -> Critique -> Implementation handoff sequence and name the weakest link.

04

Produce the artifact and include the evidence that proves it: A UI control-surface critique sheet with context inputs, artifact visibility, review criteria, and implementation handoff.

Put it into practice

Give this grounded prompt to Codex or Claude after watching.

You are helping me turn one specific YouTube video into real, durable learning.

Source video:
- Title: Google's NEW EGM Skill Set + Deepseek is ACTUALLY INSANE!
- URL: https://www.youtube.com/watch?v=jyz4ALAahdE
- Topic: Interfaces + Open Design
- My current learning frame: Choose a non-trivial feature, brainstorm it into a concept brief, draft it section by section, run the five-stage fresh-context review, resolve the findings, and require a human approval decision before implementation begins.
- Why this matters: New playlist item from AICodeKing; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:19 / Evidence 1: "twice is where it runs. It works in Antigravity, which is Google's own agent. But, it also has first-class install instructions for Claude Code, Codex, Claude Cowork, and even ChatGPT work. And because it's all just skills and..."
- 3:17 / Evidence 2: "go. Now, let's talk about the workflow. In Claude code, all the commands use the /egm: prefix. The first one is /egm:egm brainstorm. You give it a feature, something like add webhook delivery to this project. The agent..."
- 5:34 / Evidence 3: "the Claude code version has its own adapters. Each reviewer gets read-only tools and its own isolated context, so the goldfish actually stays a goldfish. There are a couple of extra skills as well. {slash} egm {colon} egm..."
- 7:09 / Evidence 4: "flash and the new DeepSeek V4.1 flash alongside GLM 5.3, GLM 5.3 flash and QN3 Mac Max. That's a really strong set of models for less than a single month of most premium plans. Now, the reason this..."
- 8:53 / Evidence 5: "and Klein can do exactly the same thing. So, clone the whole repo, not just the skills folder, point Klein at the skill you want, and let it spawn fresh sub-agents for the goldfish. Just make sure the..."
- 11:08 / Evidence 6: "long agent sessions. If you are building anything non-trivial, I'd say give the brainstorm and review steps a try on your next feature and see if the goldfish catch something you missed. And if you want to do..."

Video-aware target:
- Prompt lane: Coding-agent workflow
- Mechanism to extract: Find the workflow rule that explains when and how to use Codex, Claude Code, browser control, dashboards, or manual review.
- Artifact to produce: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
- Artifact must include: task class; agent/tool choice; context packet; verification step; handoff/recovery rule

Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Find the workflow rule that explains when and how to use Codex, Claude Code, browser control, dashboards, or manual review. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step
   - answers to these source questions: What workflow pain is being solved? | What exact coordination mechanism is shown? | What evidence proves it changes the work?
   - 3 concrete examples that apply the video idea to real agentic work, such as rate-limit routing; browser verification after a UI edit; long-running /goal session review
   - 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: choosing tools by hype; losing context across agents; letting parallel sessions become invisible
   - a checklist for the next real workflow, focused on: routing decision, context portability, verification, handoff summary
   - one practical exercise with a clear done signal: Route three recent tasks across Codex, Claude, browser checks, and manual review with a reason for each.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.

Quality bar:
- Make this specific to "Google's NEW EGM Skill Set + Deepseek is ACTUALLY INSANE!", not a generic Interfaces + Open Design essay.
- Each workflow rule must point to a timestamped claim or demo moment, then state what remains unproven.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic Codex vs Claude comparison; feature lists without task routing; claims that ignore limits or recovery.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.

Misconceptions

What to stop believing.

A beautiful page is automatically a good learning tool.

Learning requires sequence, active recall, feedback, and application.

Generated UI should be accepted as-is.

Generated UI needs critique, revision, and browser verification.

Practice studio

Learning only counts when you make something.

01

Transcript evidence map

Separate what the video actually says from what you already believe about the topic.

3 source-backed takeaways with timestamps, confidence, and a transfer note.
02

One useful artifact

Apply the video to a real workflow and produce a coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal..

A reusable artifact with a done signal and one verification step.
03

Coding-agent workflow teach-back card

Explain the coding-agent workflow mechanism to someone who has not watched the video yet.

A 90-second explanation, one diagram, one example, and one misconception to avoid.

Recall check

Answer first, then reveal — without rewatching.

How do the elephant, goldfish, and human divide responsibility?

What sequence takes an EGM feature from an idea to an approved design?

When should EGM be skipped, and what does its stronger review cost?

Source shelf

Use the video as a doorway, then verify with primary sources.

ReadingOpen Design Repogithub.com/open-design-dev/open-designReadingReact Docsreact.dev/