More screenshots
Works with the AI Agents you already use
AI Infrastructure, Costs and Security
Provision GPUs, run private AI, control cost and protect access. 10 ordered skills, starting with Cloud Cost Engineering (FinOps), FinOps for AI, Secrets and Identity Management.
$4
AI Infrastructure, Costs and Security
Example session with this skill installed
We are spending $15k/month on OpenAI and it is growing. We use LiteLLM as a gateway. Help me set up a budget and attribution plan to find out which team is burning the credits.
- Read your context and instructions
- Compiled the ai infrastructure costs
I have analyzed your request for AI API cost monitoring. Based on your use of LiteLLM, here is a structured plan to implement attribution and budgets.
AI API Cost Monitoring & Attribution Plan
| Component | Action | Tooling/Config |
|---|---|---|
| Attribution | Map one API key per team/environment | LiteLLM Key Management |
| Global Budget | Set hard monthly limit of $20,000 | litellm_settings: {max_budget: 20000} |
| Team Budgets | Implement $2,000 soft alerts per team | LiteLLM Team table in DB |
| Rate Limiting | Limit tokens per minute (TPM) by tier | LiteLLM tpm_limit per key |
| Detection | Alert on 300% spikes vs 7-day baseline | Webhook to Slack/PagerDuty |
Proposed LiteLLM Configuration
model_list:
- model_name: gpt-4
litellm_params:
model: openai/gpt-4
api_key: os.environ/OPENAI_API_KEY
general_settings:
master_key: sk-1234
database_url: postgresql://user:pass@localhost:5432/db
# Attribution logic
allow_user_auth: true
budget_manager: true
Risk Assessment
- Critical: Currently, all spend is unattributed. You cannot enforce "pay-per-use" without key rotation.
- Warning: LiteLLM budgets require a persistent database (Postgres/Redis) to track state across restarts.
- Defense: Enabling
max_parallel_requestswill prevent "denial-of-wallet" if an agent enters an infinite loop.
Next steps
- Connect LiteLLM to a Postgres instance to enable persistent budget tracking.
- Issue unique keys to the Marketing and Engineering teams to begin data collection.
- Review the first week of logs to set realistic "soft" alert thresholds.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Platform and FinOps leaders often need a repeatable process, not a one-off prompt. This pack targets: Provision GPUs, run private AI, control cost and protect access.
What's inside (10 skills, suggested learning order)
- Cloud Cost Engineering (FinOps) — Optimize cloud and GPU spend.
- FinOps for AI — Control inference and training cost.
- Secrets and Identity Management — Control credentials and access.
- Zero Trust Architecture — Security architecture without implicit trust.
- GPU Cloud and Accelerator Provisioning — Provision GPUs for AI.
- On-Premises/Private AI Deployment — Run AI in a private environment for compliance.
- Sovereign AI Infrastructure — Deploy AI with data residency.
- Technical Debt Management — Prioritize and reduce technical debt.
- IoT and Sensor Integration — Connect devices and telemetry.
- AI API Cost and Usage Monitoring — Budgets, alerts and API usage limits.
Why a skill pack instead of a one-off prompt
Each skill gives the agent a defined trigger, step-by-step procedure and output format, so results are consistent across sessions and easier to review.
How to use
- Install the pack in your SKILL.md-compatible agent.
- Ask for the task in plain language.
- Review the output against the skill's quality checks before using it.
Known limitations
- Outputs are drafts that need human review; they are not professional, legal or financial advice.
- Tools, APIs and prices change; check current vendor documentation before relying on version-specific details.
- Quality depends on the context you give the agent (data, requirements, constraints).
- Tested only where stated in the Compatibility note.
Also by Rodrigo Santos
Specialized Agents and Distribution; AI Agents: Fundamentals and Protocols; Model Fine-Tuning and Alignment
Requires manual input of billing exports or architecture specs. Does not perform live API calls to providers or execute code. Not a substitute for legal counsel or industrial safety certification.
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 any payment.
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 3 days ago