The Future of AI Agents Just Arrived ( /goal for Claude Code & Codex)
Treat `/goal` as an agentic completion contract: state the desired outcome, define proof of done, and let the agent continue through planning, execution, and verification.
Jay E | RoboNuggets14 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.
This is a concise operating pattern for moving from prompt-response work to outcome-driven agent sessions.
Skill you build: Setting up autonomous long-running agent sessions by writing a strong definition of done and acceptance criteria so /goal knows precisely when to stop.
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.
3,022 cleaned transcript words reviewed across 806 timed caption segments.
Thesis
The Future of AI Agents Just Arrived ( /goal for Claude Code & Codex) teaches a practical coding-agent workflow move: Treat `/goal` as an agentic completion contract: state the desired outcome, define proof of done, and let the agent continue through planning, execution, and verification.
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.
1:14
Definition of done
“actually introduced by Codex just 2 weeks ago. It got really popular over at X and now the Entropic team pretty much just copied it from Codeex and I think that's fine because that just gives us more...”
The /goal feature is the Ralph loop built into the harness: you set a completion condition and the agent keeps reworking each turn until that condition is satisfied, so the quality of your definition of done determines whether autopilot succeeds. Write a /goal prompt for a small task and explicitly state a concrete, checkable definition of done (e.g. 'loads with no manifest errors and shortcuts work') rather than leaving it implicit.
6:42
Haiku self-check loop
“weekly rate limits reset, you won't have any regrets with this huge portion of your tokens being unutilized. But now, let's actually use gold for a more complex task. And what we'll do is do that for both...”
At the end of each turn Claude Code invokes the lightweight Haiku model to assess whether your definition of done was met; if not, it loops again, which is also why you can frontload token-intensive batch work before a weekly rate-limit reset. Plan a batch task (like generating multiple newsletter drafts from a skill) and run it with /goal to consume otherwise-wasted tokens before your limit resets.
9:46
Plan-driven goal runs
“a lot to be desired when it comes to the design and the visuals of this. And interestingly, Claude code when it described the different AI labs in here, it described entropic as honest, helpful, and harmless. Google...”
For complex tasks, first generate a markdown plan (stack, in/out of scope, constraints, definition of done, acceptance criteria, turn budget) and point /goal at it; the turn budget can be set to unlimited and acceptance criteria are what the model checks against to decide when to stop. Generate a 'plan for goal' markdown file for a non-trivial build, then refine the acceptance criteria (e.g. forcing generated sprites to actually be used in the final HTML) before executing it.
01
Inspect context
Start with this video's job: Treat `/goal` as an agentic completion contract: state the desired outcome, define proof of done, and let the agent continue through planning, execution, and verification. Treat "Inspect context" as the outcome you are trying to make visible, not a topic label. Anchor it to 1:14, where the video says: “actually introduced by Codex just 2 weeks ago. It got really popular over at X and now the Entropic team pretty much just copied it from Codeex and I think that's fine because that just gives us more...”
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 6:42, where the video says: “weekly rate limits reset, you won't have any regrets with this huge portion of your tokens being unutilized. But now, let's actually use gold for a more complex task. And what we'll do is do that for both...”
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.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: Treat `/goal` as an agentic completion contract: state the desired outcome, define proof of done, and let the agent continue through planning, execution, and verification.
02
Explain the practical stakes without hype: This is a concise operating pattern for moving from prompt-response work to outcome-driven agent sessions.
03
Map the idea onto the Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
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: The Future of AI Agents Just Arrived ( /goal for Claude Code & Codex)
- URL: https://www.youtube.com/watch?v=aEDq1bBynOg
- Topic: Agentic Engineering
- My current learning frame: Use /goal in Claude Code to one-shot a single-file browser game from a markdown plan, deliberately sharpening the acceptance criteria so the agent reuses its own generated assets and self-verifies before stopping.
- Why this matters: This is a concise operating pattern for moving from prompt-response work to outcome-driven agent sessions.
Transcript anchors from this exact video:
- 1:14 / Evidence 1: "actually introduced by Codex just 2 weeks ago. It got really popular over at X and now the Entropic team pretty much just copied it from Codeex and I think that's fine because that just gives us more..."
- 3:00 / Evidence 2: "Cloud Code will do is try to complete that task and then check itself at the end of that turn at the end of that particular loop if it satisfied this definition of done that you gave it."
- 4:39 / Evidence 3: "use goal in order to batch create let's say articles or content or other regular automations that you need to be running the following week and just do it today before your weekly rate limits reset using the..."
- 6:42 / Evidence 4: "weekly rate limits reset, you won't have any regrets with this huge portion of your tokens being unutilized. But now, let's actually use gold for a more complex task. And what we'll do is do that for both..."
- 9:46 / Evidence 5: "a lot to be desired when it comes to the design and the visuals of this. And interestingly, Claude code when it described the different AI labs in here, it described entropic as honest, helpful, and harmless. Google..."
- 11:23 / Evidence 6: "house to pick. So, let's say we want to play as entropic. And there you go. It has much nizer visuals because of that GPT image 2 capability. And let me just zoom out here so we can..."
- 12:56 / Evidence 7: "order to oneshot as much as possible a project like this then what I would do is take a lot of time to just refine this definition of done and this acceptance criteria because that is what the..."
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 "The Future of AI Agents Just Arrived ( /goal for Claude Code & Codex)", not a generic Agentic Engineering 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.
Agentic engineering means letting agents do everything.
It means designing work so agents can do bounded pieces well.
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.
What does the /goal feature do under the hood after each turn, and which model does Claude Code use to do it?
Why does the presenter stress writing a concrete definition of done, and what example did he give for the Chrome extension task?
For a complex /goal run, what should you generate first, and which two fields in it most directly control how the run proceeds and stops?
Source shelf
Use the video as a doorway, then verify with primary sources.
