Add /gpu-ops skill — GPU infrastructure + per-agent LLM FinOps

[VibOpsai]

/gpu-ops

Deploy LLMs, track GPU cost per AI agent, enforce budgets and model policies — via VibOps MCP (74 tools, MIT).

What it does:

• GPU fleet management (deploy, scale, monitor across 6 vendors)
• Per-agent GPU cost attribution via transparent LLM proxy
• Budget enforcement per agent (HTTP 429 at limit)
• Model access policies per agent (HTTP 403 if denied)
• Dependency graph + impact analysis
• Immutable audit chain (HMAC-SHA256)

Example:

You: Which agent costs the most in GPU this month?
Claude: get_agent_usage → supply-chain-optimizer $4,559 (53%)

You: Set a $1,500 budget on marketing
Claude: set_agent_budget("marketing", 1500) → done, currently at 76%

Install: pip install git+https://github.com/VibOpsai/vibops-mcp.git
License: MIT

[trunk-io]

Merging to main in this repository is managed by Trunk.

• To merge this pull request, check the box to the left or comment /trunk merge below.

After your PR is submitted to the merge queue, this comment will be automatically updated with its status. If the PR fails, failure details will also be posted here