A comprehensive utility for tracking, analyzing, and optimizing AI API expenditures across major LLM providers to minimize unnecessary spending.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cm-api-cost-tracker
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cm-api-cost-tracker using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is API Cost Tracker?
The API Cost Tracker is a robust tool designed to provide granular visibility into your AI infrastructure spending. By parsing billing exports, usage logs, or direct API responses from providers like OpenAI, Anthropic, OpenRouter, and Google AI, it empowers developers to understand exactly where their budget is going. This utility is a core part of a professional Openclaw Skills setup, enabling teams to move beyond high-level billing summaries into detailed unit-cost analysis.
Beyond simple tracking, the tool identifies optimization opportunities by suggesting model downgrades, identifying caching potential, and highlighting prompt compression candidates. It serves as a financial audit layer for AI agents, ensuring that performance remains high while costs are kept under strict control through automated reporting and trend detection.
API Cost Tracker Use Cases
- Monitoring daily spending across multiple LLM providers in a single dashboard.
- Identifying the most expensive individual API requests or specific sessions.
- Auditing token usage to find opportunities for prompt engineering optimizations.
- Setting and tracking budget thresholds to prevent unexpected monthly billing spikes.
- Comparing current model costs against cheaper alternatives using benchmark-based quality analysis.
How API Cost Tracker Works
- The script ingests usage data from a local file export, a live API call, or environment variables.
- It normalizes disparate data formats from various providers into a unified schema containing timestamps, models, and token counts.
- The engine applies the internal pricing database to calculate or verify accurate costs per request.
- Data is aggregated and grouped based on the requested dimension, such as by model type, time period, or specific feature tags.
- An optimization engine runs a comparison between current usage and potential savings from alternative models.
- A formatted report is generated in the user's preferred format, such as a terminal table or a Markdown document.
API Cost Tracker Setup
The API Cost Tracker requires Python 3.8 or higher and operates using only the standard library, requiring no external dependencies. To start using this with your Openclaw Skills, follow these steps:
# Analyze an OpenRouter activity export
python3 scripts/api_cost_tracker.py openrouter --file activity.json
# Analyze OpenAI usage from an environment key
export OPENAI_API_KEY='your_key_here'
python3 scripts/api_cost_tracker.py auto --days 30
API Cost Tracker Data Schema & Taxonomy
The tool normalizes all incoming provider data into a consistent internal structure to ensure accurate cross-platform analysis:
| Attribute | Description |
|---|---|
| timestamp | The exact time the API request was initiated |
| model | The specific model identifier (e.g., gpt-4o, claude-3-5-sonnet) |
| tokens_in | Total number of input or prompt tokens |
| tokens_out | Total number of output or completion tokens |
| cost | The final calculated cost in USD based on the pricing database |
| metadata | Optional tags for feature-based or session-based tracking |
API Cost Tracker Advanced Features
- Trend detection and anomaly flagging to alert users to unusual spending velocity.
- Automated budget alerts that provide projected monthly overrun warnings based on current usage.
- Model substitution analysis to simulate savings if switching to a more cost-effective provider.
- Support for custom pricing databases to account for enterprise discounts or specific contract rates.
- Multi-format output support including JSON and CSV for integration into larger data pipelines within Openclaw Skills workflows.
SKILL.md
Loading
Related Openclaw Skills
Skylv Metacognition Engine
A framework that enables AI agents to monitor their own reasoning, detect biases, and improve decision quality through structured self-examination loops.
Paper to Table
A high-fidelity extraction tool designed to transform unstructured academic papers into organized literature review tables.
SEO Agent
A professional SEO optimization assistant providing real-time keyword research, content audits, and rank tracking using live API data.
Alibaba Cloud DDoS Pro Intercept Query
A specialized diagnostic tool for identifying, analyzing, and remediating request blocking reasons within Alibaba Cloud DDoS Pro security environments.
Secrets Scanner
A specialized security tool designed to identify and flag hardcoded secrets, API keys, and sensitive credentials within your source code.
Media Content Generator
Generate high-quality promotional videos and media content from images, audio, and clips using an AI-powered cloud rendering pipeline.
Featured*
