Build Your Own Coding Agent Like Pi (With 1 Prompt)
Owain demos Neo, his own Go-based coding agent, then builds a working coding agent from scratch in a single main.go file by feeding an architecture doc to an existing agent, breaking it into tasks, and shipping the core loop against Open Router. The point is that Claude Code, Pi, and Codex are all the same thing underneath: an LLM invoked in a loop with tools, roughly a thousand lines before polish.
Owain Lewis19 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 Owain Lewis; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to build and reason about a coding-agent harness from first principles, implementing the LLM loop, tool-call execution, and event hooks yourself so you can extend or swap any part of it instead of being limited by someone else's harness.
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.
4,300 cleaned transcript words reviewed across 1,194 timed caption segments.
Thesis
Build Your Own Coding Agent Like Pi (With 1 Prompt) teaches a practical coding-agent workflow move: Owain demos Neo, his own Go-based coding agent, then builds a working coding agent from scratch in a single main.go file by feeding an architecture doc to an existing agent, breaking it into tasks, and shipping the core loop against Open Router. The point is that Claude Code, Pi, and Codex are all the same thing underneath: an LLM invoked in a loop with tools, roughly a thousand lines before polish.
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.
0:20
Workflows beat one-shots
“using a single prompt. Once you own the harness, you can build any features or extensions you want. I'll link all of the resources and everything you need to get started in the description below. So, let's get...”
Neo turns a request into a visible multi-step workflow rather than a single prompt, and runs tools in parallel (five at once in the demo) plus parallel subagents. His task-to-PR skill encodes the workflow every developer already follows: check out a branch, read the ticket, plan, change code, test, review, address findings, publish, wait for build checks, hand back to a human, with the agent even commenting back on review feedback so you know it was addressed. Write out the exact steps you personally take from ticket to merged PR, then encode that sequence as a reusable skill or command instead of re-prompting it every time.
7:11
Architecture, then tasks
“code to make sure we're using this model when we build it. So this entire architecture doc was generated through a skill. I've built these coding agents quite a few times. So what you can do is just...”
The build starts with an architecture doc generated by a skill, read into the agent's context, then a plan skill breaks it into about eight tasks: Go foundation and main.go, types, checks and tests, the Open Router provider, the agent loop, exact-match file editing, and bounded command execution. He notes the tradeoff he deliberately accepts here, that dumping all tasks from a markdown file into one context window is worse than working one task at a time, which is why he normally tracks them as GitHub issues. Before your next build, write the architecture doc first, have the agent decompose it into a numbered task list, and run the tasks one at a time in separate sessions so each stays reviewable.
12:30
The loop is the agent
“but now that you have the foundation in place, you have everything that you need to start building on this coding agent. You can start adding more logic to this coding agent. You can improve the the build...”
The harness sends a system prompt, user message, and tool definitions to the model, which either returns a response or requests tool calls; the harness executes those calls (the LLM cannot touch your machine) and feeds results back, looping until the model stops. Emitting events rather than inlining logic like permission checks keeps that loop clean, and that event stream is exactly what a Claude Code hook listens to. Sketch the loop for your own harness on one page (request, tool-call branch, execute, feed back, terminate) and mark where you would emit events for permissions, logging, and hooks before writing any code.
01
Inspect context
Start with this video's job: Owain demos Neo, his own Go-based coding agent, then builds a working coding agent from scratch in a single main.go file by feeding an architecture doc to an existing agent, breaking it into tasks, and shipping the core loop against Open Router. The point is that Claude Code, Pi, and Codex are all the same thing underneath: an LLM invoked in a loop with tools, roughly a thousand lines before polish. Treat "Inspect context" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:20, where the video says: “using a single prompt. Once you own the harness, you can build any features or extensions you want. I'll link all of the resources and everything you need to get started in the description below. So, let's get...”
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 7:11, where the video says: “code to make sure we're using this model when we build it. So this entire architecture doc was generated through a skill. I've built these coding agents quite a few times. So what you can do is just...”
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: Owain demos Neo, his own Go-based coding agent, then builds a working coding agent from scratch in a single main.go file by feeding an architecture doc to an existing agent, breaking it into tasks, and shipping the core loop against Open Router. The point is that Claude Code, Pi, and Codex are all the same thing underneath: an LLM invoked in a loop with tools, roughly a thousand lines before polish.
02
Explain the practical stakes without hype: New playlist item from Owain Lewis; 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: Build Your Own Coding Agent Like Pi (With 1 Prompt)
- URL: https://www.youtube.com/watch?v=QER-0DaC-Gk
- Topic: Creative Automation
- My current learning frame: Write an architecture doc for a minimal coding agent, have an existing agent decompose and build it into one file against Open Router with read, edit, and bash tools, then use that agent to add its own next feature such as an AGENTS.md loader or a permissions hook.
- Why this matters: New playlist item from Owain Lewis; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:20 / Evidence 1: "using a single prompt. Once you own the harness, you can build any features or extensions you want. I'll link all of the resources and everything you need to get started in the description below. So, let's get..."
- 2:18 / Evidence 2: "can build work and workflows and automations and you can dispatch all of your coding tasks to a bunch of different coding agents. You can also manage multiple different Git repositories at the same time. So this project..."
- 4:25 / Evidence 3: "lot of these standards and processes, it's really easy to scale your workflow. So, we're just waiting for the agent to run through and basically react to some of these comments. So, I have an AI automated code..."
- 7:11 / Evidence 4: "code to make sure we're using this model when we build it. So this entire architecture doc was generated through a skill. I've built these coding agents quite a few times. So what you can do is just..."
- 10:43 / Evidence 5: "then run through. One of the reasons I don't like using markdown files for this is we're adding a lot to the agent's context window. Essentially, we're asking it to to look at all of the tasks at..."
- 12:30 / Evidence 6: "but now that you have the foundation in place, you have everything that you need to start building on this coding agent. You can start adding more logic to this coding agent. You can improve the the build..."
- 17:03 / Evidence 7: "thousand lines of code, but it's actually a capable, workable coding agent. At this point, there's obviously a lot of things we're missing here. We're missing things like context management. We're missing a whole bunch of other features..."
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 "Build Your Own Coding Agent Like Pi (With 1 Prompt)", not a generic Creative Automation 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.
Creative AI removes the need for taste.
It increases the need for taste because output volume explodes.
The best prompt is enough.
References, critique, iteration, and post-production matter just as much.
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 does the task-to-PR workflow actually automate?
Why does he prefer GitHub issues over a tasks.md file for the task breakdown?
What is the core loop of every coding agent, and why must the harness run the tools?
Source shelf
Use the video as a doorway, then verify with primary sources.
