Getting AI to work in complex codebases (github.com)

Interesting read, and some interesting ideas, but there's a problem with statements like these:

> Sean proposes that in the AI future, the specs will become the real code. That in two years, you'll be opening python files in your IDE with about the same frequency that, today, you might open up a hex editor to read assembly.

> It was uncomfortable at first. I had to learn to let go of reading every line of PR code. I still read the tests pretty carefully, but the specs became our source of truth for what was being built and why.

This doesn't make sense as long as LLMs are non-deterministic. The prompt could be perfect, but there's no way to guarantee that the LLM will turn it into a reasonable implementation.

With compilers, I don't need to crack open a hex editor on every build to check the assembly. The compiler is deterministic and well-understood, not to mention well-tested. Even if there's a bug in it, the bug will be deterministic and debuggable. LLMs are neither.

I've used this pattern on two separate codebases. One was ~500k LOC apache airflow monolith repo (I am a data engineer). The other was a greenfield flutter side project (I don't know dart, flutter, or really much of anything regarding mobile development).

All I know is that it works. On the greenfield project the code is simple enough to mostly just run `/create_plan` and skip research altogether. You still get the benefit of the agents and everything.

The key is really truly reviewing the documents that the AI spits out. Ask yourself if it covered the edge cases that you're worried about or if it truly picked the right tech for the job. For instance, did it break out of your sqlite pattern and suggest using postgres or something like that. These are very simple checks that you can spot in an instant. Usually chatting with the agent after the plan is created is enough to REPL-edit the plan directly with claude code while it's got it all in context.

At my day job I've got to use github copilot, so I had to tweak the prompts a bit, but the intentional compaction between steps still happens, just not quite as efficiently because copilot doesn't support sub-agents in the same way as claude code. However, I am still able to keep productivity up.

-------

A personal aside.

Immediately before AI assisted coding really took off, I started to feel really depressed that my job was turning into a really boring thing for me. Everything just felt like such a chore. The death by a million paper cuts is real in a large codebase with the interplay and idiosyncrasies of multiple repos, teams, personalities, etc. The main benefit of AI assisted coding for me personally seems to be smoothing over those paper cuts.

I derive pleasure from building things that work. Every little thing that held up that ultimate goal was sucking the pleasure out of the activity that I spent most of my day trying to do. I am much happier now having impressed myself with what I can build if I stick to it.

The fundamental frustration most engineers have with AI coding is that they are used to the act of _writing_ code being expensive, and the accumulation of _understanding_ happening for free during the former. AI makes the code free, but the understanding part is just as expensive as it always was (although, maybe the 'research' technique can help here).

But let's assume you're much better than average at understanding code by reviewing it -- you have another frustrating experience to get through with AI. Pre-AI, let's say 4 days of the week are spend writing new code, while 1 day is spent fixing unforseen issues (perhaps incorrect assumption) that came up after production integration or showing things to real users. Post-AI, someone might be able to write those 4 days worth of code in 1 day, but making decisions about unexpected issues after integration doesn't get compressed -- that still takes 1 day.

So post-AI, your time switches almost entirely from the fun, creative act of writing code to the more frustrating experience of figuring out what's wrong with a lot of code that is almost correct. But you're way ahead -- you've tested your assumptions much faster, but unfortunately that means nearly all of your time will now be spent in a state of feeling dumb and trying to figure out why your assumptions are wrong. If your assumptions were right, you'd just move forward without noticing.

I built a package which I use for large codebase work[0].

It starts with /feature, and takes a description. Then it analyzes the codebase and asks questions.

Once I’ve answered questions, it writes a plan in markdown. There will be 8-10 markdowns files with descriptions of what it wants to do and full code samples.

Then it does a “code critic” step where it looks for errors. Importantly, this code critic is wrong about 60% of the time. I review its critique and erase a bunch of dumb issues it’s invented.

By that point, I have a concise folder of changes along with my original description, and it’s been checked over. Then all I do is say “go” to Claude Code and it’s off to the races doing each specific task.

This helps it keep from going off the rails, and I’m usually confident that the changes it made were the changes I wanted.

I use this workflow a few times per day for all the bigger tasks and then use regular Claude code when I can be pretty specific about what I want done. It’s proven to be a pretty efficient workflow.

[0] GitHub.com/iambateman/speedrun

This article is like a bookmark in time of where I exactly gave up (in July) managing context in Claude code.

I made specs for every part of the code in a separate folder and that had in it logs on every feature I worked on. It was an API server in python with many services like accounts, notifications, subscriptions etc.

It got to the point where managing context became extremely challenging. Claude would not be able to determine business logic properly and it can get complex. e.g. if you want to do a simple RBAC system with an account and profile with a junction table for roles joining an account with profile. In the end what kind of worked was I had to give it UML diagrams of the relationship with examples to make it understand and behave better.

It's strange that author is bragging that this 35K LOC was researched and implemented in 7 hours, but there are 40 commits spanning across 7 days. Was it 1 hour per day or what?

Also quite funny that one of the latest commits is "ignore some tests" :D

There are a lot of people declaring this, proclaiming that about working with AI, but nobody presents the details. Talk is cheap, show me the prompts. What will be useful is to check in all the prompts along with code. Every commit generated by AI should include a prompt log recording all the prompts that led to the change. One should be able to walkthrough the prompt log just as they may go through the commit log and observe firsthand how the code was developed.

So I can attest to the fact that all of the things proposed in this article actually works. And you can try it out yourself on any arbitrary code base within few minutes.

This is how: I work for a company called NonBioS.ai - we already implement most of what is mentioned in this article. Actually we implemented this about 6 months back and what we have now is an advanced version of the same flow. Every user in NonBioS gets a full linux VM with root access. You can ask nonbios to pull in your source code and ask it to implement any feature. The context is all managed automatically through a process we call "Strategic Forgetting" which is in someways an advanced version of the logic in this article.

Strategic Forgetting handles the context automatically - think of it like automatic compaction. It evaluates information retention based on several key factors:

1. Relevance Scoring: We assess how directly information contributes to the current objective vs. being tangential noise

2. Temporal Decay: Information gets weighted by recency and frequency of use - rarely accessed context naturally fades

3. Retrievability: If data can be easily reconstructed from system state or documentation, it's a candidate for pruning

4. Source Priority: User-provided context gets higher retention weight than inferred or generated content

The algorithm runs continuously during coding sessions, creating a dynamic "working memory" that stays lean and focused. Think of it like how you naturally filter out background conversations to focus on what matters.

And we have tried it out in very complex code bases and it works pretty well. Once you know how well it works, you will not have a hard time believing that the days of using IDE's to edit code is probably numbered.

Also - you can try it out for yourself very quickly at NonBioS.ai. We have a very generous free tier that will be enough for the biggest code base you can throw at nonbios. However, big feature implementations or larger refactorings might take time longer than what is afforded in the free tier.

A few weeks later, @hellovai and I paired on shipping 35k LOC to BAML, adding cancellation support and WASM compilation - features the team estimated would take a senior engineer 3-5 days each.

Sorry, had they effectively estimated that an engineer should produce 4-6KLOC per day (that's before genAI)?

if you haven't tried the research -> plan -> implementation approach here, you are missing out on how good LLMs are. it completely changed my perspective.

the key part was really just explicitly thinking about different levels of abstraction at different levels of vibecoding. I was doing it before, but not explicitly in discrete steps and that was where i got into messes. The prior approach made check pointing / reverting very difficult.

When i think of everything in phases, i do similar stuff w/ my git commits at "phase" levels, which makes design decision easier to make.

I also do spend ~4-5 hours cleaning up the code at the very very end once everything works. But its still way faster than writing hard features myself.

It seems we're still collectively trying to figure out the boundaries of "delegation" versus "abstraction" which I personally don't think are the same thing, though they are certainly related and if you squint a bit you can easily argue for one or the other in many situations.

> We've gotten claude code to handle 300k LOC Rust codebases, ship a week's worth of work in a day, and maintain code quality that passes expert review.

This seems more like delegation just like if one delegated a coding task to another engineer and reviewed it.

> That in two years, you'll be opening python files in your IDE with about the same frequency that, today, you might open up a hex editor to read assembly (which, for most of us, is never).

This seems more like abstraction just like if one considers Python a sort of higher level layer above C and C a higher level layer above Assembly, except now the language is English.

Can it really be both?

Good pointers on decompositing and looking at implementation or fixing in chunks.

1. Break down the feature or bug report into a technical implementation spec. Add in COT for the splits. 2. Verify the implementation spec. Feed reviews back to your original agent that has created the spec. Edit, merge, integrate feedback. 3. Transform implementation spec into an implementation plan - logically split into modules look at dependency chain. 4. Build, test and integrate continuously with coding agents 5. Squash the commits if needed into a single one for the whole feature.

Generally has worked well as a process when working on a complex feature. You can add in HITL at each stage if you need more verification.

For larger codebases always maintain an ARCHITECTURE.md and for larger modules a DESIGN.md

> It was uncomfortable at first. I had to learn to let go of reading every line of PR code. I still read the tests pretty carefully, but the specs became our source of truth for what was being built and why.

This is exactly right. Our role is shifting from writing implementation details to defining and verifying behavior.

I recently needed to add recursive uploads to a complex S3-to-SFTP Python operator that had a dozen path manipulation flags. My process was:

* Extract the existing behavior into a clear spec (i.e., get the unit tests passing).

* Expand that spec to cover the new recursive functionality.

* Hand the problem and the tests to a coding agent.

I quickly realized I didn't need to understand the old code at all. My entire focus was on whether the new code was faithful to the spec. This is the future: our value will be in demonstrating correctness through verification, while the code itself becomes an implementation detail handled by an agent.

> Within an hour or so, I had a PR fixing a bug which was approved by the maintainer the next morning

An hour for 14 lines of code. Not sure how this shows any productivity gain from AI. It's clear that it's not the code writing that is the bottleneck in a task like this.

Looking at the "30K lines" features, the majority of the 30K lines are either auto-generated code (not by AI), or documentation. One of them is also a PoC and not merged...

Maybe I am just misunderstanding. I probably am; seems like it happens more and more often these days

But.. I hate this. I hate the idea of learning to manage the machine's context to do work. This reads like a lecture in an MBA class about managing certain types of engineers, not like an engineering doc.

Never have I wanted to manage people. And never have I even considered my job would be to find the optimum path to the machine writing my code.

Maybe firmware is special (I write firmware)... I doubt it. We have a cursor subscription and are expected to use it on production codebases. Business leaders are pushing it HARD. To be a leader in my job, I don't need to know algorithms, design patterns, C, make, how to debug, how to work with memory mapped io, what wear leveling is, etc.. I need to know 'compaction' and 'context engineering'

I feel like a ship corker inspecting a riveted hull

> Heck even Amjad was on a lenny's podcast 9 months ago talking about how PMs use Replit agent to prototype new stuff and then they hand it off to engineers to implement for production.

Please kill me now

Nice read and I'm trying to follow and use your tools. But it's just seems hard to make Claude Code follow the instructions - it always diverges into researching and proposing fixes as well as searching for root causes - which is strictly said it the research_codebase not to do.

From many tries I had a success only twice. Could you give some hints on how to use it?

Context has never been the bottleneck for me. AI just stops working when I reach certain things that AI doesn't know how to do.

1. Research -> Plan -> Implement

2. Write down the principles and assumptions behind the design and keep them current

In other words, the same thing successful human teams on complex projects do! Have we become so addicted to “attention-deficit agile” that this seems like a new technique?

Imagine, detailed specs, design documents, and RFC reviews are becoming the new hotness. Who would have thought??

I use a similar pattern but without the subagents. I get good results with it. I review and hand edit "research" and plans. I follow up and hand edit code changes. It makes me faster, especially in unfamiliar codebases.

But the write up troubles me. If I'm reading correctly, he did 1 bugfix (approved and merged) and then 2 larger PRs (1 merged, 1 still in draft over a month later). That's an insanely small sample size to draw conclusions from.

How can you talk like you've just proven the workflow works "for brownfield codebases"? You proved it worked for 2/3 tasks in 2 codebases, one failure (we can't say it works until the code is shipped IMO).

I am still sceptical of the roi and the time i am supposed to sink into trying and learning these AI tools which seem to be replacing each other every week.

For me the biggest difficulty is I find it hard to read unverifiable documentation. It's like dyslexia - if I can't connect the text content with runnable code, I feel lost in 5 minutes.

So with this approach of spending 3 hours on planning without verification in code, that's too hard for me.

I agree the context compaction sounds good. But I'm not sure if an md file is good enough to carry the info from research to plan and implementation. Personally I often find the context is too complex or the problem is too big. I just open a new session to resolve a smaller, more specific problem in source code, then test and review the source code.

Except for ofc pushing their own product (humanlayer) and some very complex prompt template+agent setups that are probably overkill for most, the basics in this post about compaction and doing human review at the correct level are pretty good pointers. And giving a bit of a framework to think within is also neat

> And yeah sure, let's try to spend as many tokens as possible

It'd be nice if the article included the cost for each project. A 35k LOC change in a 350k codebase with a bunch of back and forth and context rewriting over 7 hours, would that be a regular subscription, max subscription, or would that not even cover it?

Thanks to write such detailed article... lot of very well supported information.

I've been working on something what I call Micromanaged Driven Development https://mmdd.dev and wrote about it at https://builder.aws.com/content/2y6nQgj1FVuaJIn9rFLThIslwaJ/...

I'm in a similar search and I'm stoked to see that many people riding the wave of coding with AI is moving in this direction.

Lots of learning ahead.

This article bases its argument on the predicate that AI _at worst_ will increase developer productivity be 0-10%. But several studies have found that not to be true at all. AI can, and does, make some people less effective

I used a similar pattern. When ask AI to do a large implementation. I ask gemini-2.5-pro to write a very detailed overview implementation plan. Then review it. Then ask gemini-2.5-pro to split the plan into multiple stages and write detail implementation plan for each stage. Then I ask claude sonnat to read the overview plan and implement the stage n. I found that this is the only way to complete a major implementation with a relatively high success rate.

Can't agree with the formula for performance, on the "/ size" part. You can have a huge codebase, but if the complexity goes up with size then you are screwed. Wouldn't a huge but simple codebase be practical and fine for AI to deal with?

The hierarchy of leverage concept is great! Love it. (Can't say I like the 1 bad line of CLAUDE.md is 100K lines of bad code; I've had some bad lines in my CLAUDE.md from time to time - I almost always let Claude write it's own CLAUDE.md.).

Tasted like a sales pitch the whole way and what do ya know at the very end there it is

> Sean proposes that in the AI future, the specs will become the real code. That in two years, you'll be opening python files in your IDE with about the same frequency that, today, you might open up a hex editor to read assembly (which, for most of us, is never).

Only if AI code generation is correct 99.9% of the time and almost never hallucinates. We trust compilers and don't read assembly code because we know it's deterministic and the output can never be wrong (barring bugs and certain optimization issues, which are rare/one-time fixes). As long as generated code is not doing what the original "code" (in this case, specs) doing, humans need to go back to fix things themselves.

Hello, I noticed your privacy policy is a black page with text seemingly set to 1% or so opacity. Can you get the slopless AI to fix that when time permits?

- Mr. Snarky

I enjoyed the emphasis on optimising the context window itself. I think that's the most important bit.

An abstraction for this that seems promising to me for its completeness and size is a User Story paired with a research plan(?).

This works well for many kinds of applications and emphasizes shipping concrete business value for every unit of work.

I wrote about some of it here: https://blog.nilenso.com/blog/2025/09/15/ai-unit-of-work/

I also think a lot of coding benchmarks and perhaps even RL environments are not accounting for the messy back and forth of real world software development, which is why there's always a gap between the promise and reality.

I wrote this blogpost on the same topic: https://getstream.io/blog/cursor-ai-large-projects/

It's super effective with the right guardrails and docs. It also works better on languages like Go instead of Python.

I'm using GPT Pro and a VS extension that makes it easy to copy code from multiple files at once. I'm architecting the new version of our SaaS and using it to generate everything for me on the backend. It’s a huge help with modeling and coding, though it takes a lot of steering and correction. I think I’ll end up with a better result than if I did it alone, since it knows many patterns and details I’m not aware of (even simple things like RRULE). I’m designing this new project with a simpler, more vertical architecture in the hopes that Codex will be able to create new tables and services easily once the initial structure is ready and well documented.

Edit: typo.

As an aside, this single markdown file as an entire GitHub repo is a unique approach to blog posts.

Re the meta of running multiple phases of "document expansion":

Research helps with complex implementations and for brownfield. But it isn't always needed - simple bugfixes can be one-shot!

So all AI workflows could be expressed with some number "N" of "document expansion phases":

N(0): vibe coding.

N(1): "write a spec then implement it while I watch".

N(2): "research then specify". At this point you start to get serious steerability.

What's N(3) and beyond? Strategy docs, industry research, monetization planning? Can AI do these too, all of it ending up in git? Interesting to muse on.

Verifying behavior is great and all if you can actually exhaustively test the behaviors of your system. If you can't, then not knowing what your code is actually doing is going to set you back when things do go belly up.

I used to do these things manually in Cursor. Then I had to take a few months off programming, and when I came back and updated Cursor I found out that it now automatically does ToDos, as well as keeps track of the context size and compresses it automatically by summarising the history when it reaches some threshold.

With this I find that most of the shenanigans of manual context window managing with putting things in markdown files is kind of unnecessary.

You still need to make it plan things, as well as guide the research it does to make sure it gets enough useful info into the context window, but in general it now seems to me like it does a really good job with preserving the information. This is with Sonnet 4

YMMV

I’m not an expert in either language, but seeing a 20k LoC PR go up (linked in the article) would be an instant “lgtm, asshole” kind of review.

> I had to learn to let go of reading every line of PR code

Ah. And I’m over here struggling to get my teammates to read lines that aren’t in the PR.

Ah well, if this stuff works out it’ll be commoditized like the author said and I’ll catch up later. Hard to evaluate the article given the authors financial interest in this succeeding and my lack of domain expertise.

I created an account to say this: RepoPrompt's 'Context Builder' feature helps a ton with scoping context before you touch any code.

It's kind of like if you could chat with Repomix or Gitingest so they only pull the most relevant parts of your codebase into a prompt for planning, etc

I'm a paying RepoPrompt user but not associated in any other way.

I've used it in conjunction with Codex, Claude Code, and any other code gen tool I have tried so far. It saves a lot of tokens and time (and headaches)

It’s refreshing to read a full article this was written by a human. Content +++

To minimise context bloat and provide more holistic context, I extract on first step the important elements from a codebase via AST which then the LLM uses to determine which files to get in full for given task.

https://github.com/piqoni/vogte

I am working on a project with ~200k LoC, entirely written with AI codegen.

These days I use Codex, with GPT-5-Codex + $200 Pro subscription. I code all day every day and haven't yet seen a single rate limiting issue.

We've come a long way. Just 3-4 months ago, LLMs would start doing a huge mess when faced with a large codebase. They would have massive problems with files with +1k LoC (I know, files should never grow this big).

Until recently, I had to religiously provide the right context to the model to get good results. Codex does not need it anymore.

Heck, even UI seems to be a solved problem now with shadcn/ui + MCP.

My personal workflow when building bigger new features:

1. Describe problem with lots of details (often recording 20-60 mins of voice, transcribe)

2. Prompt the model to create a PRD

3. CHECK the PRD, improve and enrich it - this can take hours

4. Actually have the AI agent generate the code and lots of tests

5. Use AI code review tools like CodeRabbit, or recently the /review function of Codex, iterate a few times

6. Check and verify manually - often times, there are a few minor bugs still in the implementation, but can be fixed quickly - sometimes I just create a list of what I found and pass it for improving

With this workflow, I am getting extraordinary results.

AMA.

The problem is the research phase will fail because you can't glean tribal product knowledge from just looking at the code

Using AI to help with code felt like working with a smart but slightly unreliable teammate. If I wasn’t clear, it just couldn’t follow. But once I learned to explain what I wanted clearly and specifically, it actually saved me time and helped me think more clearly too.

> context management, and keeping utilization in the 40%-60% range (depends on complexity of the problem).

Is this a rule of thumb? Will the cheaper (fewer params) models dumb down at 25%?

Nice, my experience writing production code in large codebase here, granted it's evolved a lot since: https://philippcannons.com/100x-ai-coding-for-real-work-summ...

Not surprisingly, building really fast is not the silver bullet you'd think it is. It's all about what to build and how to distribute it. Otherwise bigcos/billionaires would have armies of engineers growing their net worth to epic scales.

How granular are the specs? Is it at the level of "this is the code you must write, and here is how to do it", or are you letting AI work some of that out?

Thanks for sharing, I wonder how do you keep the stylistic and mental alignment of the codebase - is this happens during the code review or there are specific instructions during at the plan/implement stages?

> our team of three is averaging about $12k on opus per month

That’s usd 150k per year. Probably low for SF, but may be a lot in other areas.

Doesnt githubs new speckit solve this? https://github.com/github/spec-kit

Honestly I was reading that article and I smelled sales pitch and then in the end of course.

This is not my experience at all.

I also don’t get the line obsession.

Good code has less lines not more

Lots of gold in this article. It's like discovering a basket of cheat codes. This will age well.

Great links, BAML is a crazy rabbithole and just found myself nodding along to frequent /compact. These tips are hard-earned and very generously given. Anyone here can take it or leave it. I have theft on my mind, personally. (ʃƪ¬‿¬)

When I read about people dumping 2000 lines of code every few days, I'm extremely skeptical about the quality of this code. All the people I've met who worked at this rate were always going for naive solutions and their code was full of hard-to-see bugs which only reared their ugly heads once in a while and were impossible to debug.

We're currently in a transition phase where we're using agentic coding on systems developed with tools and languages designed for humans. Ironically, this makes things unnecessarily hard as things that are easy for us aren't necessary easy to deal with; or that optimal for agentic coding systems.

People like languages that are expressive and concise. That means they do things like omit types, use type inference, macros, syntactic sugar, allow for ambiguities and all the other stuff that gives us shorter, easier to type code that requires more effort to figure out. A good intuition here might be that the harder the compiler/interpreter has to work to convert it into running/executable code, the harder an LLM will have to work to figure out what that code does.

LLMs don't mind verbosity and spelling things out. Things that are long winded and boring to us are helpful for an LLM. The optimal language for an LLM is going to be different than one that is optimal for a human. And we're not good at actually producing detailed specifications. Programming actually is the job of coming up with detailed specifications. Easy to forget when you are doing that but that's literally what programming is. You write some kind of specification that is then "compiled" into something that actually works as specified.

The solution to agentic coding isn't writing specifications for our specifications. That just moves the problem.

We've had a few decades of practice where we just happen to stuff code into files and use very primitive tools to manipulate those files. Agentic coding uses a few party tricks involving command line tools to manipulate those files and reading them by one into the precious context window. We're probably shoveling too much data around. But since that's the way we store code, there are no better tools to do that.

From having used things like Codex, 99% of what it does is interrogating what's there via tediously slow prodding and poking around the code base using simple command line commands and build tool invocations. It's like watching paint dry. I usually just go off doing something else while it boils the oceans and does god knows what before finally doing the (usually) relatively straightforward thing that I asked it to do. It's easy to see that this doesn't scale that well.

The whole point of a large code base is that it probably won't all fit in the context window. We can try to brute force the problem; or we can try to be more selective. The name of the game here is being able to be able to quickly select just the right stuff to put in there and discard all the rest.

We can either do that manually (tedious and a lot of work, sort of as the article proposes), or make it easier for the LLM to use tools that do that. Possibly a bunch of poorly structured files in some nested directory hierarchy isn't the optimal thing here. Most non AI based automated refactorings require something that more closely resembles the internal data structures of what a compiler would use (e.g. symbol tables, definitions, etc.).

A lot of what an agentic coding system has to do is reconstruct something similar enough to that just so it can build a context in which it can do constructive things. The less ambiguous and more structured that is, the easier the job. The easier we make it to do that, the more it can focus on solving interesting problems rather than getting ready to do that.

I don't have all the answers here but if agentic coding is going to be most of the coding, it makes sense to optimize the tools, languages, etc. for that rather than for us.

[dead]

Why though. Why should we do that?

If AI is so groundbreaking, why do we have to have guides and jump through 3000 hoops just so we can make it work?

[dead]

[dead]

TLDR:

We're taking a profession that attracts people who enjoy a particular type of mental stimulation, and transforming it into something that most members of the profession just fundamentally do not enjoy.

If you're a business leader wondering why AI hasn't super charged your company's productivity, it's at least partly because you're asking people to change the way they work so drastically, that they no longer derive intrinsic motivation from it.

Doesn't apply to every developer. But it's a lot.

[dead]

[dead]

[dead]

[dead]

I refactored CPython using GPT-5, turning the compiler bilingual for english and portuguese keywords.

https://github.com/ricardoborges/cpython

what web programming task GPT-5 can't handle?

OpenAI Codex has an `update_plan` function[0]. I'm wondering if switching the implementation to this would improve the coding agent's capabilities or is the default for simplicity better.

[0]: https://blog.toolkami.com/openai-codex-tools/