agentsview
Local tool for browsing AI coding agent sessions, transcripts, and token usage.
Recommended for its usage and token-cost metrics123 and searchable transcripts across most coding agents,456 built on local conversation files instead of a proxy.78 The only reported problem is one user who couldn't get filtering by agent to work9 and saw cost figures that differed from ccusage.10
Why people recommend it
Usage metrics · 3
Shows token usage, visualizations, and what usage would cost at API rates.
“I can highly recommend https://agentsview.io/ for a way to dig into your LLM usage.”
“use one of the many token/session recording tools (I like agentsview)”
Show 1 more
“I recommend https://github.com/wesm/agentsview - its quite good, has usage metrics and nice viewer/search for transcripts from most agents.”
Transcript browsing · 2
Lets users search transcripts and inspect each conversation, including tool calls.
“I can highly recommend https://agentsview.io/ for a way to dig into your LLM usage.”
“I recommend https://github.com/wesm/agentsview - its quite good, has usage metrics and nice viewer/search for transcripts from most agents.”
Multi-agent coverage · 2
Shows activity across all or most coding agents in one place.
“This is the best one I found”
“I recommend https://github.com/wesm/agentsview - its quite good, has usage metrics and nice viewer/search for transcripts from most agents.”
No proxy · 2
Reads local conversation files, so it doesn't add a proxy as another point of failure.
“This is the best one I found”
“I can highly recommend https://agentsview.io/ for a way to dig into your LLM usage.”
Works well · 2
Users call it the best option they found and say it works well.
“This is the best one I found”
“I've been running https://github.com/kenn-io/agentsview for this, works well.”
Caveats and dissent
Agent filtering · 1
One user couldn't get the config change or --agent flag to filter results as expected.
“Hmm, I tried that and made the config file change but it didn't work for me.”
Cost accuracy · 1
One user saw cost figures that differed completely from ccusage and couldn't tell which was right.
“Hmm, I tried that and made the config file change but it didn't work for me.”
Threads where it came up
- If you don't run your own LLM proxy, how do you know whats going on? 1 mention
- Show HN: Frugal Tokens – explore costs and usage across coding agents 1 mention
- Mozilla: The state of open source AI 1 mention
- Show HN: ctx – Search the coding agent history already on your machine 1 mention
- Claude Fable 5 1 mention
- Paseo – Beautiful open-source coding agent interface (desktop, mobile, CLI) 1 mention
Data: agentsview.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 6
“This is the best one I found”
“I can highly recommend https://agentsview.io/ for a way to dig into your LLM usage.”
“use one of the many token/session recording tools (I like agentsview)”
“I've been running https://github.com/kenn-io/agentsview for this, works well.”
“Hmm, I tried that and made the config file change but it didn't work for me.”
“I recommend https://github.com/wesm/agentsview - its quite good, has usage metrics and nice viewer/search for transcripts from most agents.”