oh-my-hi plugin

A Claude Code plugin with /omh: a local dashboard of skills, agents, MCP, token use and cache for Claude Code and Codex.

  • 60 ★ on the repository
  • Last commit
  • MIT
Plugin

Extended access

Why this level

  • The plugin reads local agent history and configuration and starts a local HTTP server for the dashboard.
All reasons and checks

netil/oh-my-hi

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add plugins/oh-my-hi

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • The plugin reads local agent history and configuration and starts a local HTTP server for the dashboard.

This is third-party code. Review the repository files before installing.

What it does

oh-my-hi parses the Claude Code and Codex configuration and usage data, stores the result in a local SQLite database and serves an interactive dashboard through a local HTTP server. The `/omh` command builds the dashboard and opens it in the browser. It shows skills, agents, plugins, hooks, memory, MCP servers, rules and commands, plus an estimate of how context splits between visible and invisible. Separate screens cover tokens: breakdown by model and day, the contribution of each skill, agent and MCP server, the effect of a short cache TTL and a month-end cost projection against a budget. The context window explorer replays a past session turn by turn and heatmaps show which skills, agents and commands you call most.

Who it is for. People who spend long hours in Claude Code and Codex and want to see which skills, agents and MCP servers eat context and money.

Good fit when

  • The token bill grew and you need the skills and agents with the largest share.
  • You suspect a short cache TTL wastes money: the panel counts losses by day.
  • You want to replay an old session turn by turn and see how the context window filled.
  • You need Claude Code and Codex side by side by tokens, cost and structure.

Not a fit when

  • You need central team analytics: data stays on your computer.
  • You use another agent: only Claude Code and Codex are supported.
  • You need exact vendor invoices: cost is computed from local data and tariffs.

Example request

Run /omh and show which skills and MCP servers spent the most tokens over the last week.

Limitations

The dashboard reads local configuration and history files of both tools, so it sees only what sits on this computer. Cost and context estimates are calculated and depend on the tariffs built into the project. Each tool gets its own prices and wording. Builds run through the `bin/oh-my-hi.mjs` script, and the `/omh --update` and `/omh --status` flags handle updates and auto rebuild state.

How to disable. Remove the plugin with `claude plugin uninstall oh-my-hi`. The local SQLite database and built dashboard files can be deleted separately.

FAQ

How do I start the dashboard?

In Claude Code run `/omh`: the plugin parses the data, builds the dashboard and opens it. For specific projects pass a path: `/omh <path>`.

What is Token Breakdown?

A breakdown of use by context type: skill, agent, MCP server and built-in tool. It shows the cost of startup context multiplied by session count and a top five per type.

How is the cache TTL impact found?

The panel detects waste from a short cache lifetime: best and worst seven day efficiency, an estimate of extra spend and a daily hit rate trend.

Can I look at Codex and Claude Code together?

Yes. Pick a tool from the selector: each has its own tokens, cost, sessions and structure with prices and wording specific to it.

How do I update the plugin?

With `/omh --update`: it checks and installs the latest version. `/omh --status` shows the auto rebuild state.

Discussion

No questions or reviews yet.

Plugins

A dashboard for every Claude Code session on the machine: live agents, cost, history search and MCP tools for Claude.

PluginExtended access126Repository stars

ClawMetry reads the session files agents already write and shows the timeline, tool calls, tokens and cost in one dashboard.

CLIExtended access436Repository stars

A local real-time dashboard for Claude Code and Codex sessions: agent activity and tool calls

CLIExtended access1.2KRepository stars

A local dashboard for Claude Code sessions: timelines, token costs, tools, agents, skills and live sessions

CLIExtended access339Repository stars

A terminal and web dashboard for Claude Code sessions, costs and settings, with budget alerts and a 30-day forecast.

CLIExtended access101Repository stars

A local dashboard for Claude Code: a live graph of sessions and subagents, tokens, cost and file conflicts.

PluginExtended access104Repository stars

Comparison with similar tools

Cards that share the main search query, and how they differ.

ToolOfficialRepository starsMain difference
oh-my-hi (this card)no60
AI Observerno292a local panel of tokens, cost, latency and errors for several agents; oh-my-hi also shows structure: skills, agents, MCP and startup context cost
Claude Session Dashboardno77a dashboard over the `.claude` folder with subagent chains; oh-my-hi adds Codex, cache and cost projection

README in short

The project collects data from the configuration and logs of two coding agents into SQLite and presents it as a dashboard with use and context estimates. It supports multi-tool and multi-workspace views with global and per-project scope.

README badge

Report a problem