- Home
- Skills
- Agents & Orchestration
- AI agent usage track
More screenshots
Works with the AI Agents you already use
AI agent usage track
A Herdr side panel mod to track Claude Code, Cursor, Copilot, and Agy quotas, costs, and context in real-time.
$4.99
AI agent usage track
Example session with this skill installed
Install the agent-usage mod for Herdr and set up my Claude Code status line tracking.
- Read your context and instructions
- Compiled the ai agent usage
I have prepared the Herdr agent-usage mod files. I will now deploy the capture scripts and update your Claude Code settings.
Installation Summary
| Action | Target Path / Setting | Status |
|---|---|---|
| Create Directory | ~/.config/herdr-app/mods/agent-usage/ | Success |
| Copy Mod Files | mod/* | Completed |
| Script Deploy | claude-statusline.sh | Verified |
| Script Deploy | agy-statusline.sh | Skipped (No Agy detected) |
| Patch Settings | ~/.claude/settings.json | Updated |
| System Check | herdr-app check | Passed |
I have updated your ~/.claude/settings.json to include the new statusLine command. This script will save status JSON to ~/.cache/herdr-agent-usage/claude/ every 5 seconds without interfering with your CLI output.
"statusLine": {
"type": "command",
"command": "bash ~/.config/herdr-app/mods/agent-usage/scripts/claude-statusline.sh",
"refreshInterval": 5
}
Next steps
- Restart your Claude Code session.
- Send one message in Claude Code to initialize the data file.
- Open the Herdr side panel to view your new Usage cards.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing AI subscriptions across multiple CLI tools is a fragmented mess. This skill installs the agent-usage mod for Herdr, centralizing quota tracking, spend monitoring, and active agent context in a single 280px sidebar. It solves the problem of hitting unexpected rate limits or losing track of model costs across different environments.
What it does
- Quota visualization Tracks Claude Code, Codex, and Agy usage with 5-hour and 7-day bars including reset times.
- Spend tracking Monitors Cursor model requests and total spend over a rolling 30-day window.
- Active agent monitoring Displays real-time cost, model type, and context window usage for every running agent.
- Privacy-first architecture Reads local JSON logs only. Makes zero network calls and requires no API keys or credentials.
- Customizable UI Supports provider card reordering via drag-and-drop and selective visibility toggles.
How it works
- Initialize scripts The skill reads the status-line bash scripts and prepares the local cache directories.
- Deploy mod folder Files are copied to the Herdr configuration directory.
- Configure CLI tools The skill updates
~/.claude/settings.jsonand Agy settings to pipe status data to the mod. - Validation It runs a system check via the Herdr CLI to ensure the mod is correctly recognized.
Frameworks & tools
This skill interfaces with Herdr, Claude Code, Cursor, Copilot CLI, and Agy (Antigravity CLI). It uses Bash scripts for status line capture and local JSON files for data persistence.
Why this beats prompting it yourself
Manual setup of cross-tool status line piping is error-prone and requires specific shell handling to avoid blocking the agent's main output. This skill automates the file system scaffolding, JSON patching, and validation steps required for a functional dashboard.
Use cases
- Prevent Claude Code rate limits by monitoring 5-hour and 7-day quota bars.
- Audit Cursor monthly spend by tracking specific model request volumes.
- Debug context window overflow by viewing real-time usage bars for active agents.
- Centralize disparate CLI metrics into a single desktop GUI panel.
Known limitations
Requires Herdr to be installed. Users must restart their agents and send one message to trigger the initial data population. Does not support non-local data sources.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Requires Herdr. Agents must be restarted after install. Usage data only populates after the first message is sent to each agent. Status line scripts require a Bash-compatible environment.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago