AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Observability/
Claude-Code-Agent-Monitor
Claude-Code-Agent-Monitor logo

Claude-Code-Agent-Monitor

Active·★ 1.2k·MIT·Updated 2026-11-12

### Architecture & Positioning Claude-Code-Agent-Monitor (CCAM) is a local-first observability dashboard and session tracking system engineered specifically for terminal AI coding agents, including Claude Code, Cursor, and Codex. When agents execute long-running autonomous workflows across multi-file edits, compiler tests, and bash scripts, tracking context bloat, sub-agent spawning, and tool-call latency inside standard CLI scrolling becomes nearly impossible. CCAM resolves this observability deficit by passively discovering local JSONL session transcripts and hook events, rendering a unified, real-time control plane in a browser dashboard or native macOS app. ### Core Capabilities & Differentiation CCAM's defining advantage lies in its multi-provider aggregation and hierarchical task deconstruction. Rather than displaying an undifferentiated stream of tool outputs, CCAM reconstructs conversations into intuitive parent-child agent trees, detailing sub-agent delegations, tool arguments, and execution runtimes. Developers gain full-text search across historical transcripts, live token-to-cost conversion estimators, and interactive task controls that let users inspect, resume, or replay paused agent trajectories directly from the UI. ### Deployment & Best Practices To set up CCAM, install the package via Node.js (22+) and configure provider hooks in your shell configuration or local agent settings so session logs are mirrored to the monitored path. Because CCAM is strictly local-first, no code snippets, prompts, or sensitive environment secrets are ever transmitted to third-party tracking servers, making it compliant for enterprise proprietary codebases. Keep the dashboard pinned to a secondary monitor or terminal split pane to maintain constant visibility over long-running coding agents.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-11-14.

License
MIT
Latest release
v2.4.6 · 2026-11-12
Last commit
2026-11-12
Contributors
40
Forks
266
Open issues and PRs
45
First published
2026-03-05

Source: GitHub, npm and PyPI public APIs.

#ai-agents#claude-agents#claude-code#claude-skills#dev-tooling#express#expressjs#macos-app
$ Install
$ git clone https://github.com/hoangsonww/Claude-Code-Agent-Monitor.git && cd Claude-Code-Agent-Monitor && npm run setup
↗ Visit site★ GitHub
01

Features

01Combines provider sessions in shared dashboard views.
02Shows conversations, tool calls and parent-child hierarchies.
03Tracks usage and API-equivalent cost estimates.
02

Why choose it

+Provides a unified, real-time local dashboard aggregating sessions across Claude Code, Cursor, and Codex.
+Reconstructs linear tool call streams into intuitive parent-child hierarchical task trees with sub-agent visibility.
+Tracks real-time token consumption and translates usage into accurate API-equivalent cost estimations.
+100% local-first architecture ensures proprietary code and private project transcripts never leave the developer machine.
03

Trade-offs

!Requires initial configuration of shell hooks or provider environment paths to enable passive transcript discovery.
!Parsing extremely large, long-running agent JSONL session transcripts can cause momentary local indexing lag.
!Task resumption controls depend on downstream CLI tool state compatibility and may fail if workspace git state has diverged.
!Mobile or remote web access requires manual configuration of reverse proxies or local port forwarding.
04

Maintenance check

ActiveChecked: 2026-11-13
Last commit
2026-11-12
Commits, last 90 days
100+
Releases, last 90 days
22
Latest release
v2.4.6 (2026-11-12)
Open issues
44
Recent issues with at least one comment
1/5
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 11, 32, 20, 28, 15, 54, 15, 60, 17, 19, 22, 1

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #376 [Feature]: Support Google Antigravity CLI (agy) and Hermes Agent sessions2 comments · 2026-10-17
  • #197 i18n: community translations for new languages (Spanish, German, French, ...) -- contributors welcome2 comments · 2026-06-29
  • #315 [Feature]: Trim hook payloads before storing events.data (whole-file mirrors grew my DB to 2 GB)1 comments · 2026-09-03
  • #305 [Feature]: Tiered retention with rollup-before-delete so history shrinks instead of disappearing1 comments · 2026-08-24
  • #149 Feature: Local-first backup, restore, and sync workflow1 comments · 2026-05-12

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

nodejs
Node.js 22.22+ and npm 9+
Verified via docs
browser
Local web dashboard; selected-provider hooks need setup
Verified via docs
06

Quick start

1
$ git clone https://github.com/hoangsonww/Claude-Code-Agent-Monitor.git
2
$ cd Claude-Code-Agent-Monitor
3
$ npm run setup
07

Use cases

↳Inspect activity across several coding agents.
↳Review tool-flow and concurrency during a task.
↳Resume a local agent run from the dashboard.
08

Alternatives

⚡Top VS:
VSnanoclaw★31.1k→VSdarwin-skill★7.0k→VStaste-skill★117.7k→VSnuwa-skill★38.5k→
nanoclaw logo
nanoclaw★ 31.1k
A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK
VSCompare→
video-shotcraft logo
video-shotcraft★ 21.0k
AI video skill for Claude Code & Codex — cinematic, editable product promos with Remotion: 124 shot recipe cards, 150 motion previews, beat sync, sound design, Motion Workbench & JianYing (CapCut CN) export.
claude-skills logo
claude-skills★ 12.4k
67 Specialized Skills for Full-Stack Developers. Transform Claude Code into your expert pair programmer.
darwin-skill logo
darwin-skill★ 7.0k
达尔文.skill —— 一个让你的Skill无限进化的系统:评估→改进→测试→保留或回滚 | Autoresearch-inspired autonomous skill optimization for Claude Code. Evaluate, improve, test, keep or revert.
VSCompare→
Papermorph logo
Papermorph★ 1.8k
An AI skill that turns books into animated, narrated, interactive web experiences
orca logo
orca★ 127.8k
Orca lets you run a fleet of coding agents (e.g. Claude Code, Codex) with your own subscription. Available on desktop, mobile and remote runtime.
taste-skill logo
taste-skill★ 117.7k
Agent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
VSCompare→
nuwa-skill logo
nuwa-skill★ 38.5k
你想蒸馏的下一个员工,何必是同事。蒸馏任何人的思维方式——心智模型、决策启发式、表达DNA。Distill how anyone thinks.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
hoangsonww /
Claude-Code-Agent-Monitor
↗

Local-first observability dashboard and session tracker for Claude Code, Cursor, and Codex with transcript visualization, token cost estimation, and task resumption

1.2k266MIT
Last commit6d ago
StatusActive
CategoryObservability
Contributors40
Latest releasev2.4.6
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.