This looks useful. With AI generated code the hardest part is reviewing it.
A normal git diff gets messy once the agent changes several files for different reasons. Grouping the change into “chapters” seems like the right idea.
Do you infer those chapters only from the diff, or can you also use the agent’s original plan/task history?
Minor nitpick: This isn't what I expected when I read "CLI". I envisioned a terminal-native experience. Unless I skimmed over this way too fast, this is a browser experience that you trigger from the terminal.
EDIT: I should mention that I think the idea is cool. We're in a new age where reviewing large amounts of unfamiliar code has become a larger problem than it was previously.
I mean it’s quite literally a command line interface to their tool… what else should it be called that differentiates it from a pure browser flow?
What you are describing sounds more like “TUI” than “CLI” imo. A CLI is an interface—it’s about the input step. It makes no promise about what happens after that.
I've been spending a lot of my energy lately on how to run eng teams where we:
1. Maximize long-term shipping velocity
2. Maximize quality (whatever that means)
3. Maintain minimal complexity
4. Are intentional about which skills we let atrophy, which we keep sharp, and which new ones we have to build
5. Make juniors more capable, not just more productive
These are always in tension.
I've been thinking about instituting some sort of socratic method during planning and review plus spaced interval testing to ensure both the humans and AI coding agents understand and find some max of the factors above.
And yeah, I think number 5 on your list is particularly interesting - juniors will develop much slower if they don't go through the struggle of understanding implementation
We're hoping that our tool can help make that easier
Looks cool! Chapters is definitely something I've been angling towards as well. Any plans on going in the other direction (directly incorporating rich feedback/review into the agent loop through Stage)?
appreciate it! and yep, we've got lots of ideas on the roadmap to bring a more complete iteration experience closer to the coding agent.
we've found it pretty silly that we have to push to GitHub in order to get comments from a review bot, pull them down locally, then rinse and repeat. the whole agentic coding landscape could benefit from some centralization
yeah, i definitely feel like we're currently in a very time-sparse model for review, when a lot of changes can be condensed locally. it'd reduce a lot of friction and also save a lot of compute costs if we were able to left-shift a lot of our current review work
love this.
i had the same issue with ai generated code and wrote parley. https://parley.cloudflavor.io
it's a TUI that can help you review code by enabling you to comment on the diff itself.
but i like this approach of organizing code into chapters. i think what my tool is missing this exact thing.
parley looks awesome! we're planning to add support for comments soon, which is definitely a key feature to being able to iterate back and forth with a coding agent
Interested to try this! Have you thought about separating the parts of a PR that are routine/uninteresting from the parts that are load-bearing and need more careful review?
yes! we've found that reading changes this way make it very easy to separate the important stuff from the unimportant stuff, and we're thinking of ways to make that more visible in the UI
Cool, good to hear. I think it’s often the case even within an individual file or change that it’s 90% routine and 10% critical to review. That’s a big part of the problem in my mind.
This looks useful. With AI generated code the hardest part is reviewing it.
A normal git diff gets messy once the agent changes several files for different reasons. Grouping the change into “chapters” seems like the right idea.
Do you infer those chapters only from the diff, or can you also use the agent’s original plan/task history?
the cool part is that you can run the skill from the same agent session so it has the context on the plan and implementation process
but if you run the skill in a fresh session, it naturally wouldn't have the plan
Minor nitpick: This isn't what I expected when I read "CLI". I envisioned a terminal-native experience. Unless I skimmed over this way too fast, this is a browser experience that you trigger from the terminal.
EDIT: I should mention that I think the idea is cool. We're in a new age where reviewing large amounts of unfamiliar code has become a larger problem than it was previously.
yep sorry about that, we weren't exactly sure what the best framing was
glad you like the idea though! let us know what you think
I mean it’s quite literally a command line interface to their tool… what else should it be called that differentiates it from a pure browser flow?
What you are describing sounds more like “TUI” than “CLI” imo. A CLI is an interface—it’s about the input step. It makes no promise about what happens after that.
While you are not factually incorrect, my expectations were subverted in the same way that OOP's were.
Looks cool and will give it a try.
I've been spending a lot of my energy lately on how to run eng teams where we:
1. Maximize long-term shipping velocity
2. Maximize quality (whatever that means)
3. Maintain minimal complexity
4. Are intentional about which skills we let atrophy, which we keep sharp, and which new ones we have to build
5. Make juniors more capable, not just more productive
These are always in tension.
I've been thinking about instituting some sort of socratic method during planning and review plus spaced interval testing to ensure both the humans and AI coding agents understand and find some max of the factors above.
Great let us know what you think!
And yeah, I think number 5 on your list is particularly interesting - juniors will develop much slower if they don't go through the struggle of understanding implementation
We're hoping that our tool can help make that easier
Looks cool! Chapters is definitely something I've been angling towards as well. Any plans on going in the other direction (directly incorporating rich feedback/review into the agent loop through Stage)?
appreciate it! and yep, we've got lots of ideas on the roadmap to bring a more complete iteration experience closer to the coding agent.
we've found it pretty silly that we have to push to GitHub in order to get comments from a review bot, pull them down locally, then rinse and repeat. the whole agentic coding landscape could benefit from some centralization
yeah, i definitely feel like we're currently in a very time-sparse model for review, when a lot of changes can be condensed locally. it'd reduce a lot of friction and also save a lot of compute costs if we were able to left-shift a lot of our current review work
love this. i had the same issue with ai generated code and wrote parley. https://parley.cloudflavor.io it's a TUI that can help you review code by enabling you to comment on the diff itself. but i like this approach of organizing code into chapters. i think what my tool is missing this exact thing.
parley looks awesome! we're planning to add support for comments soon, which is definitely a key feature to being able to iterate back and forth with a coding agent
Love the idea. This would have been a game-changer in previous projects I've worked on.
> We’ve found that reading changes this way is a lot easier for us than reading them in an IDE or other similar CLI tools
If this tool was in the terminal I'd use it.
we're planning on adding it!
This feature exists already. It's called git.
Interested to try this! Have you thought about separating the parts of a PR that are routine/uninteresting from the parts that are load-bearing and need more careful review?
yes! we've found that reading changes this way make it very easy to separate the important stuff from the unimportant stuff, and we're thinking of ways to make that more visible in the UI
Cool, good to hear. I think it’s often the case even within an individual file or change that it’s 90% routine and 10% critical to review. That’s a big part of the problem in my mind.