Claude Code's Team Just Dropped Their Internal Loops Guide (copy this)
This video walks through Anthropic's internal guide to how the Claude Code team uses loops, explaining their four classifications — turn-based, goal-based, time-based, and proactive — plus the guardrails that make autonomous runs safe: definitions of done, stop criteria, separate judge models, and token-cost discipline.
Mansel Scheffel10 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 Mansel Scheffel; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to choose the right loop type for a task and configure it responsibly — pairing a clear definition of done with stop criteria, an independent judge, and cost controls before letting an agent run autonomously.
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.
2,773 cleaned transcript words reviewed across 762 timed caption segments.
Thesis
Claude Code's Team Just Dropped Their Internal Loops Guide (copy this) teaches a practical coding-agent workflow move: This video walks through Anthropic's internal guide to how the Claude Code team uses loops, explaining their four classifications — turn-based, goal-based, time-based, and proactive — plus the guardrails that make autonomous runs safe: definitions of done, stop criteria, separate judge models, and token-cost discipline.
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.
0:41
Four loop levels
“going to look at a few of those ways. Anthropic classify their loops in four ways internally. They have turn-based, goal-based, time-based, and proactive. And you can see the further we go up the scale, the more autonomous...”
An agent loop is gather context, take action, verify work, repeat until done — and Anthropic classifies loops as turn-based, goal-based, time-based, and proactive, with autonomy increasing up the scale toward an 'AI employee'; every level depends on giving Claude a definition of done (often via a skill.md) so it knows what good looks like. For one task you run with Claude today, write its explicit definition of done in a skill.md-style file and identify which of the four loop levels it currently sits at.
4:48
Goals need judges
“interval and a prompt. So, the first thing that we can do is just put in, let's say, 15 minutes. And then we can also just run one of our skills instead of having to type in a...”
Goal-based loops (/goal) retry until the definition of done is met, so you must set stop criteria — a max number of tries — or it can burn tokens and API cost indefinitely, and a separate judge model must grade each turn because an agent checking its own homework can gamify the result; goals suit deterministic work like researching a lead list, not taste-based work like writing outreach DMs. Sort three of your recurring tasks into 'deterministic — safe for /goal' versus 'taste-based — keep a human in the loop', and note the retry cap you'd set for each goal candidate.
6:45
Schedules and proactive
“those separate personas all looking at that goal from different angles to make sure that it meets our definition of done. So in practice, that would look like this. We would have our forward slash schedule, every hour...”
Time-based loops via /loop take an interval plus a prompt or skill (e.g. an AI news monitor every 15 minutes) but run locally so your machine must stay on, while routines run the same job on Anthropic's cloud with plan-limited runs per day — and the proactive level chains it all: a schedule monitors Slack, feeds a goal with a skill and judge, and dynamic workflows explore three solutions in parallel worktrees with adversarial review. Set up one /loop with a sensible interval derived from how often the underlying data actually changes — not an arbitrary five minutes — and track its consumption with the usage command.
01
Inspect context
Start with this video's job: This video walks through Anthropic's internal guide to how the Claude Code team uses loops, explaining their four classifications — turn-based, goal-based, time-based, and proactive — plus the guardrails that make autonomous runs safe: definitions of done, stop criteria, separate judge models, and token-cost discipline. Treat "Inspect context" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:41, where the video says: “going to look at a few of those ways. Anthropic classify their loops in four ways internally. They have turn-based, goal-based, time-based, and proactive. And you can see the further we go up the scale, the more autonomous...”
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 4:48, where the video says: “interval and a prompt. So, the first thing that we can do is just put in, let's say, 15 minutes. And then we can also just run one of our skills instead of having to type in a...”
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: This video walks through Anthropic's internal guide to how the Claude Code team uses loops, explaining their four classifications — turn-based, goal-based, time-based, and proactive — plus the guardrails that make autonomous runs safe: definitions of done, stop criteria, separate judge models, and token-cost discipline.
02
Explain the practical stakes without hype: New playlist item from Mansel Scheffel; queued for transcript-backed review, topic mapping, and a practical learning artifact.
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: Claude Code's Team Just Dropped Their Internal Loops Guide (copy this)
- URL: https://www.youtube.com/watch?v=SRM_K2jSrgk
- Topic: Codex + Claude Workflows
- My current learning frame: Take one recurring workflow, pilot it as a single supervised run with the cheapest adequate model, then promote it to a goal-based loop with a written definition of done, a retry cap, a separate judge, and deterministic steps offloaded to scripts.
- Why this matters: New playlist item from Mansel Scheffel; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:41 / Evidence 1: "going to look at a few of those ways. Anthropic classify their loops in four ways internally. They have turn-based, goal-based, time-based, and proactive. And you can see the further we go up the scale, the more autonomous..."
- 2:32 / Evidence 2: "have a different model that checks every single turn because you absolutely do not want the same agent checking its own homework because it can cheat that or gamify it to the point where it thinks that it's..."
- 4:48 / Evidence 3: "interval and a prompt. So, the first thing that we can do is just put in, let's say, 15 minutes. And then we can also just run one of our skills instead of having to type in a..."
- 6:45 / Evidence 4: "those separate personas all looking at that goal from different angles to make sure that it meets our definition of done. So in practice, that would look like this. We would have our forward slash schedule, every hour..."
- 8:15 / Evidence 5: "these things we spoke about, we obviously have our very clear definition of done, which I'm now tired of saying, but secondly, you need to make sure you define the stop criteria. For a goal, it's the number..."
- 9:55 / Evidence 6: "here, you need to make sure that you're doing that especially when you get into running those dynamic workflows because like I said, you can run hundreds of agents and that will cost you a ridiculous amount of..."
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 "Claude Code's Team Just Dropped Their Internal Loops Guide (copy this)", not a generic Codex + Claude Workflows 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.
One agent should do every task.
Different tools have different strengths. Routing is part of the workflow.
More context is always better.
Relevant context helps; stale context causes drift and cost.
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.
What are Anthropic's four internal classifications of loops, and what changes as you move up the scale?
Why should a goal-based loop use a separate judge model instead of letting the agent verify itself?
What's the key difference between a local /loop and a cloud routine?
Source shelf
Use the video as a doorway, then verify with primary sources.
