How do I install the Openclaw Cost Tracker skill?

Run npx skillmds@latest add modbender/openclaw-cost-tracker-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Openclaw Cost Tracker safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: executes scripts, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Openclaw Cost Tracker?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Openclaw Cost Tracker free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Openclaw Cost Tracker?

modbender (@modbender) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Openclaw Cost Tracker by modbender — Agent Skill

Name: Openclaw Cost Tracker
Rating: 4.2 (5 reviews)
Author: modbender

Openclaw Cost Tracker

Track OpenClaw token usage and API costs by parsing session JSONL files. Use when user asks about token spend, API costs, model usage breakdown, daily cost trends, or wants to know how much they're spending on Claude/GPT. Shows per-model breakdown, daily spend chart, and grand totals. No API keys needed — reads directly from local session files.

modbender Updated Sep 22, 2026 12 repo stars

File contents

OpenClaw Cost Tracker

Parse OpenClaw session files to compute per-model token usage, costs, and daily spend trends. Works with any OpenClaw installation — no API keys or external services needed.

Usage

# All-time cost report
python3 scripts/cost_tracker.py

# Last 7 days
python3 scripts/cost_tracker.py --days 7

# Today only
python3 scripts/cost_tracker.py --days 1

# Since a specific date
python3 scripts/cost_tracker.py --since 2026-02-01

# JSON output for dashboards/integrations
python3 scripts/cost_tracker.py --days 30 --format json

# Custom agents directory
python3 scripts/cost_tracker.py --agents-dir /path/to/agents

What It Reports

Per-model breakdown:

Total cost, tokens, and request count
Input/output/cache token split
Visual percentage bar

Daily spend: Bar chart of cost per day (text) or structured array (JSON).

Grand totals: Combined cost, tokens, and requests across all models.

How It Works

Auto-discovers the OpenClaw agents directory (~/.openclaw/agents)
Scans all agent session JSONL files (filtered by mtime for speed)
Extracts message.usage and message.model from each entry
Aggregates by model and by day
Outputs formatted report or JSON

JSON Output Schema

{
  "models": [
    {
      "model": "claude-opus-4-6",
      "totalTokens": 220800000,
      "inputTokens": 3200,
      "outputTokens": 390800,
      "cacheReadTokens": 149400000,
      "cacheWriteTokens": 1200000,
      "totalCost": 528.55,
      "requestCount": 2088
    }
  ],
  "daily": [
    { "date": "2026-02-20", "cost": 37.14, "byModel": { "opus-4-6": 35.0, "sonnet-4": 2.14 } }
  ],
  "grandTotal": { "totalCost": 580.11, "totalTokens": 269800000, "totalRequests": 3122 },
  "meta": { "agentsDir": "...", "filesScanned": 65, "entriesParsed": 3122, "range": "7d" }
}

Integration

Feed JSON output into dashboards, alerting, or budgeting tools. The daily array is ready for charting. Set up a cron to track spend over time:

# Daily cost snapshot to file
0 0 * * * python3 /path/to/cost_tracker.py --days 1 --format json >> ~/cost-log.jsonl

Requirements

Python 3.8+
OpenClaw installed with session data in ~/.openclaw/agents/
No external dependencies (stdlib only)

---name: openclaw-cost-tracker-2description: Track OpenClaw token usage and API costs by parsing session JSONL files. Use when user asks about token spend, API costs, model usage breakdown, daily cost trends, or wants to know how much they're spending on Claude/GPT. Shows per-model breakdown, daily spend chart, and grand totals. No API keys needed — reads directly from local session files.---# OpenClaw Cost TrackerParse OpenClaw session files to compute per-model token usage, costs, and daily spend trends. Works with any OpenClaw installation — no API keys or external services needed.## Usage```bash# All-time cost reportpython3 scripts/cost_tracker.py# Last 7 dayspython3 scripts/cost_tracker.py --days 7# Today onlypython3 scripts/cost_tracker.py --days 1# Since a specific datepython3 scripts/cost_tracker.py --since 2026-02-01# JSON output for dashboards/integrationspython3 scripts/cost_tracker.py --days 30 --format json# Custom agents directorypython3 scripts/cost_tracker.py --agents-dir /path/to/agents```## What It Reports**Per-model breakdown:**- Total cost, tokens, and request count- Input/output/cache token split- Visual percentage bar**Daily spend:** Bar chart of cost per day (text) or structured array (JSON).**Grand totals:** Combined cost, tokens, and requests across all models.## How It Works1. Auto-discovers the OpenClaw agents directory (`~/.openclaw/agents`)2. Scans all agent session JSONL files (filtered by mtime for speed)3. Extracts `message.usage` and `message.model` from each entry4. Aggregates by model and by day5. Outputs formatted report or JSON## JSON Output Schema```json{ "models": [ { "model": "claude-opus-4-6", "totalTokens": 220800000, "inputTokens": 3200, "outputTokens": 390800, "cacheReadTokens": 149400000, "cacheWriteTokens": 1200000, "totalCost": 528.55, "requestCount": 2088 } ], "daily": [ { "date": "2026-02-20", "cost": 37.14, "byModel": { "opus-4-6": 35.0, "sonnet-4": 2.14 } } ], "grandTotal": { "totalCost": 580.11, "totalTokens": 269800000, "totalRequests": 3122 }, "meta": { "agentsDir": "...", "filesScanned": 65, "entriesParsed": 3122, "range": "7d" }}```## IntegrationFeed JSON output into dashboards, alerting, or budgeting tools. The `daily` array is ready for charting. Set up a cron to track spend over time:```bash# Daily cost snapshot to file0 0 * * * python3 /path/to/cost_tracker.py --days 1 --format json >> ~/cost-log.jsonl```## Requirements- Python 3.8+- OpenClaw installed with session data in `~/.openclaw/agents/`- No external dependencies (stdlib only)

Frequently asked questions