← Back to Skills Marketplace
Token Stats
by
huyiling1111
· GitHub ↗
· v2.6.1
· MIT-0
58
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install ai-agent-usage-stats
Description
选择要监控的 AI 助手 → 查看 token 消耗。支持 Hermes / Claude Code / CodeX / OpenClaw，每次都让你选
README (SKILL.md)
agent-usage-stats — 选个 Agent 看它的消耗
核心原则
- 每次运行都弹菜单 — 你想看哪个 Agent 就选哪个，不预设
- 数据来自本地 — 不联网，纯读你的 Agent 本地数据文件
- 零依赖 — 纯 Python 标准库，装完即用
前置条件
- Python 3.11+
- 至少一种 Agent 有使用记录：Hermes / Claude Code / CodeX / OpenClaw
- 安装方式：
clawhub install ai-agent-usage-stats，然后python3 ~/skills/ai-agent-usage-stats/agent-usage-stats.py setup
用法速查
# 交互式菜单（默认）
agent-usage-stats
# 跳过菜单直接看
agent-usage-stats -a hermes
agent-usage-stats -a claude-code
agent-usage-stats -a codex
# 实时监控
agent-usage-stats --watch
agent-usage-stats -a hermes --watch
# 查看本机装了哪些 Agent
agent-usage-stats --list-backends
# 查看版本
agent-usage-stats --version
Hermes 集成
Hermes 的 SKILL.md 里可以这样用：
run: agent-usage-stats -a hermes
这样每次任务结束会自动输出 Hermes 的 token 消耗，不弹菜单。
数据源说明
| Agent | 数据读哪里 |
|---|---|
| Hermes | ~/.hermes/state.db → sessions 表 |
| Claude Code | ~/.claude/projects/**/*.jsonl |
| CodeX | ~/.codex/state_*.sqlite → threads 表 |
| OpenClaw | ~/.openclaw/agents/main/sessions/ |
Usage Guidance
Install only if you are comfortable with a local tool reading your AI assistant usage/history files and changing PATH/wrapper/config files during setup or uninstall. Be especially cautious on Windows+WSL, avoid editing its saved paths.json to untrusted paths, and prefer manual ClawHub updates over the built-in update command until the WSL subprocess construction and setup/update confirmations are improved.
Capability Tags
Capability Assessment
Purpose & Capability
The core function of reading local AI-agent usage logs is coherent with the stated purpose, and the main README discloses the local data sources. However, the executable also includes setup, uninstall, and self-update flows beyond simple viewing.
Instruction Scope
Normal stats commands are user-directed and mostly scoped to selected agents, but '--all' and the menu can read multiple local agent history stores. The short SKILL.md lists fewer supported agents than the fuller README/code.
Install Mechanism
The setup flow creates wrappers and modifies PATH through shell rc files or the Windows user PATH; the update flow invokes ClawHub commands and copies updated files into the skill directory. These actions are disclosed but broad for a stats utility and lack strong confirmation/dry-run controls.
Credentials
Reading local SQLite/JSONL usage files is proportionate, but the WSL fallback dynamically builds a python3 -c script with path data, creating avoidable code-injection risk if path/config data is maliciously shaped.
Persistence & Privilege
The skill persists a wrapper in ~/.local/bin, edits user PATH configuration, stores discovered agent paths under ~/.config/agent-usage-stats, and uninstall removes those config directories. These are user-commanded, not hidden, but they materially change the local environment.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ai-agent-usage-stats - After installation, invoke the skill by name or use
/ai-agent-usage-stats - Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.6.1
Initial release as ai-agent-usage-stats
Metadata
Frequently Asked Questions
What is Token Stats?
选择要监控的 AI 助手 → 查看 token 消耗。支持 Hermes / Claude Code / CodeX / OpenClaw，每次都让你选. It is an AI Agent Skill for Claude Code / OpenClaw, with 58 downloads so far.
How do I install Token Stats?
Run "/install ai-agent-usage-stats" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Token Stats free?
Yes, Token Stats is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Token Stats support?
Token Stats is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Token Stats?
It is built and maintained by huyiling1111 (@huyiling1111); the current version is v2.6.1.
More Skills