Squad Usage — Development skill for Claude Code
Report the tokens this project's agents have used, per agent (agent-squad).
How to install Squad Usage
Installs to ~/.claude/skills/gzurl-agent-squad-squad-usage/SKILL.md
mkdir -p ~/.claude/skills/gzurl-agent-squad-squad-usage && curl -fsSL https://raw.githubusercontent.com/gzurl/agent-squad/HEAD/commands/squad-usage.md -o ~/.claude/skills/gzurl-agent-squad-squad-usage/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Squad Usage does
description: Report the tokens this project's agents have used, per agent (agent-squad) argument-hint: "[Nd | YYYY-MM-DD]"
The CEO wants to know how many tokens this project's agents have used. Your role is your session name.
**The period.** What the CEO typed after the command is: `$ARGUMENTS`.
- Nothing: the whole history.
Nd, such as30d: the last N days, today included.- A date,
YYYY-MM-DD, with or withoutsincebefore it: from that day on.
Anything else:
Alternatives in Development
- Semble — Fast and Accurate Code Search for Agents 6.1k ★
- Code Review Graph — Local knowledge graph for Claude Code 3.7k ★
- Opik Openclaw — 🦞 Official plugin for OpenClaw that exports agent traces to Opik 725 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Squad Usage All
Report the tokens every agent on this machine has used, by project; it relays nothing (agent-squad)
Verify Theme
Run the full Graphite quality gate (drift, tokens, contrast, YAML) and report pass/fail
Fable Eval
Run the fable-mode eval suite (probes → pairwise judge → report). Costs tokens — runs headless claude many tim
Mimi Logging
Show what mimi costs and saves in this project, from its Claude Code session logs, and save the report under m
Agent Report.Es
A dónde van los tokens - conversación principal frente a subagentes, por nivel de esfuerzo y por rol
Agent Report
Where the tokens go - main conversation vs subagents, by effort level and by role
Related Agents
Squad Designer
Operate as the squad's Product Designer — turn open UI/UX decisions into presentational component code, tokens
10x Tool Calls
Cursor and Windsurf meter usage by requests and tool calls rather than tokens, which means a finished or stall
Scaffold Inspector
Inspect the mvp/ folder and return a structured report on what's already scaffolded — framework, router, state