I vibe coded a Claude Code/Codex transcript viewer webapp

In the early days, when you interact with an LLM in a chat interface, you see almost everything the model sees: a set of user and model messages. The only thing hidden from the user in the UI is the system prompt. This is not the case with coding agents. The UI hides the details of tool calls, interleaved thinking traces, and subagent actions. 

To get a better sense of what my Claudes are doing, I vibe coded this transcript viewer that shows the details of all tool calls and whatnot (it also works with Codex.)

One concrete issue this tool helps expose is the insanity of Claude code’s “WebFetch” tool. Claude reads websites using the WebFetch tool, which calls a smaller LLM and returns only a tiny summary of the site. This drives me fucking insane. If you ask Claude to review a lesswrong post, it would get a tiny summary, and then use that to do the “review.” Bad Claude!

(For Codex, the web search results are hidden from the user, so I don’t even know how bad it is.)

To combat this, I also have read-lesswrong skill you can use here. The skill forces the model to read the whole post, top comments, and figures before responding. We can see in the transcript viewer that the model followed the skill instructions. Very nice.

There are also some other nice quality of life features. The viewer renders LaTeX math, shows you what subagents are doing, makes it easy to jump between user messages on the side, and has a basic search bar to look through past sessions. 

You can download the viewer here: https://github.com/tim-hua-01/cc_transcript_viewer.

Make sure to prevent transcript deletion in your Claude settings via

{

"cleanupPeriodDays": 99999,

"showThinkingSummaries": true

}

And add this:

model_reasoning_summary = "detailed"

To ~/.codex/config.toml:

I will say though, if you don’t know me personally, you should probably have a coding agent check the code for malware (or just give it the prompt request here)You should not be executing arbitrary code you saw on lesswrong lmao.