by codenotary
See what Claude Code, Codex and/or Kiro-Crew and its subagents are really doing, live in your browser or beside the chat: what it costs, how close you are to your limits, when it needs you, and a check of its work before it says done. 6 languages.
# Add to your Claude Code skills
git clone https://github.com/codenotary/agentmon-liveGuides for using ai agents skills like agentmon-live.
See how agentmon-live compares with popular alternatives.
agentmon-live is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by codenotary. See what Claude Code, Codex and/or Kiro-Crew and its subagents are really doing, live in your browser or beside the chat: what it costs, how close you are to your limits, when it needs you, and a check of its work before it says done. 6 languages. It has 51 GitHub stars.
agentmon-live's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/codenotary/agentmon-live" and add it to your Claude Code skills directory (see the Installation section above).
agentmon-live is primarily written in TypeScript. It is open-source under codenotary on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh agentmon-live against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM lists this project for discovery and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
Press o and your session opens in the browser. See who started whom, what each agent does right now, and where the money goes. See all of it.
You can use the Codex or Kiro Crew setup without running Claude Code. The browser view needs Node.js.
Using Claude Code? AgentMon also checks before it says "done". Its done gate catches unfinished work such as failed checks, stale test results and test runs that selected no tests. Changed assertions and added skips are highlighted for review.
The done gate, Focus and project learning are Claude Code features. Available monitoring data differs between tools; the installation sections explain what each integration shows.
In a Claude Code session (2.1.308 or later):
/plugin install agentmon-live --marketplace codenotary/agentmon-live
Answer y to add the marketplace and pick a scope. There is nothing to set up. The pane opens by itself when the terminal is wide enough, or with /agentmon.
claude plugin install agentmon-live --marketplace codenotary/agentmon-live
Desktop app: click + next to the prompt box, then Plugins, Add marketplace with codenotary/agentmon-live, then Add plugin.
Or paste this into a Claude Code session:
Install the Claude Code plugin agentmon-live from the marketplace codenotary/agentmon-live
(claude plugin install agentmon-live --marketplace codenotary/agentmon-live). Tell me to restart
Claude Code when done.
The web view lists every open Kiro Crew chat beside your Claude Code sessions, whatever harness the chat runs on (kiro-cli, Claude Code, Codex and the others). It reads the log Kiro Crew writes, so this needs nothing installed in Kiro Crew: open the web view and the chats are there.
To have AgentMon start with Kiro Crew and show in a tab beside the chat, install this repo as a Kiro Crew app. It needs node.
git clone https://github.com/codenotary/agentmon-live
kirocrew app install <path>/agentmon-live
Then turn it on in Kiro Crew:
Open Library in the left bar. AgentMon Live is listed under Disabled: click Enable. Kiro Crew asks you to trust the app before its code runs. That code is kiro/hooks.py, about 100 lines of standard-library Python that run with Kiro Crew's own rights and start the web view's server. Read it before you say yes.
Open a chat, click + above the side panel and pick AgentMon Live.
The tab opens on the chat beside it and follows when you switch chats; you can still pick another session in its top bar. It shows the chat live: where each turn went, tool calls, files, the session log. The tab works on Linux and macOS.
kirocrew app uninstall agentmon-live.The guide has the rest, including a registry install from Kiro Crew's settings.
The tab loads through Kiro Crew's own address, so it works wherever the Kiro Crew dashboard opens: over Tailscale, on a phone, or over SSH with only the dashboard's port forwarded (ssh -L 5476:127.0.0.1:5476 <user>@<host>). It shows the sessions to whoever Kiro Crew lets into its dashboard, invited members of a shared crew included.
The web view on its own, http://127.0.0.1:47311/, lets in only a browser of the account that runs it. From another machine, forward that port too and log in as the account that runs Kiro Crew; from any other login the page says it may not read the sessions. If Kiro Crew runs as a service of its own account, set AGENTMON_WEB_UIDS=<id -u of your login> in its environment; in a container, AGENTMON_WEB_OPEN=1 turns the check off, and any local process can then read the sessions. Restart Kiro Crew and stop a running server (pkill -f 'web.mjs serve') for either to take effect.
The web view lists every Codex chat on the machine beside your Claude Code sessions: the CLI, codex exec and the IDE extension, with the subagents each chat starts in its flow and Subagents section. It reads the session files Codex writes, so this needs nothing installed in Codex: open the web view and the chats are there. Each chat's context gauge shows Codex's own count against its own window, with each compaction in the log; it is never summed with a Claude Code session's.
With no Claude Code session open, nothing starts the web view's server. As a Codex plugin, AgentMon starts it with each Codex session. It needs node, on macOS or Linux.
codex plugin marketplace add codenotary/agentmon-live
codex plugin add agentmon-live@agentmon-live
Codex runs a plugin's hook only after you trust it: open /hooks in Codex and trust AgentMon Live's SessionStart and Stop hooks, both in codex/hooks.json. The first starts the web view's server, the second is the done gate. Then open http://127.0.0.1:47311/.
Done gate: when Codex is about to stop and a test, lint, typecheck or build failed after this turn's first code edit, or ran before its last one, the Stop hook sends Codex back once with what is open.
Status line: needs Codex 0.171 or later with [features] tui_extensions_statusline = true in ~/.codex/config.toml; then pick AgentMon Live in /statusline. The line under the prompt shows open gate items, the latest checks, context and the 5-hour limit.
Pane: a Codex build with the tui_extensions_sidebar feature shows the pane as an AgentMon tab in Codex's sidebar: open items, the running call, subagents, changes and checks, insights with a pie of where the time went, recent calls, context and limits, a session log, and w to open the web view. Stock Codex has no sidebar for plugins; use the web view.
Skill and feedback: the agentmon skill gives the web view's link; the feedback tool makes a GitHub issue link you submit yourself.
What you see: turns with their time split, tool calls with duration and result, files read and edited, prompts and answers, tokens, the context gauge and your 5-hour and weekly limits.
Verification: the last test, lint, typecheck and build command of the chat, each ok, failed or unknown.
What stays empty: approvals, subagent cards, cost in dollars and learning. The pane is part of Claude Code.
To remove it: codex plugin remove agentmon-live@agentmon-live.
The guide has the rest.
The pane needs the Claude Code terminal. The web view does not: any Claude Code run that loads the plugin shows there, such as claude -p.
claude plugin install agentmon-live --marketplace codenotary/agentmon-live.claude -p "...". The plugin starts the server and posts the session. T