Model Inventory Auditor
Inventory every LLM model and provider your code depends on, the AI bill of materials, and flag the concentration risk. When a model can be pulled offline overnight, you cannot manage a risk you cannot see.
What it does
- Lists every provider and model in your code, config, env, and manifests, with where each is used.
- Flags hardcoded model ids.
- Flags single-provider dependency with no alternative.
- Flags inconsistent references and missing config indirection.
- Prints a clear "what breaks if this provider is pulled" summary.
How to run
Point it at your repo. Read-only, no network. The provider patterns are editable.
Good to know
This is heuristic. The inventory is as complete as the provider patterns it knows, so add your own for in-house clients.
Questions
Why now?
A frontier model was pulled offline by government order three days after launch. Knowing your exposure stopped being optional.
Will it change anything?
No. It is read-only and uses no network. Provider patterns are editable so you can add your own.
See your model exposure before it bites.
Current price and purchase option are shown on Agensi. Instant download. Works with any agent that reads SKILL.md. Pairs with the Model Resilience Linter.
Get the Model Inventory Auditor on Agensi