AI Vendor Evaluator
It accompanies you to choose between several AI tools with evidence and leaves you a decision memo and the exact questions you have to ask.

Buying AI
You are going to connect your system to an AI model and no one has read the terms and conditions of that model. The question is not whether the model is good, but whether you can send it the data you plan to send it and use it for what you plan to use it for. ClariaIA already has the official documentation of a model catalog loaded, with its model card, its terms, its privacy policy and its acceptable use policy. He interviews you about your case, defines between three and six requirements that have to be met for your use to be viable, and searches for evidence page by page in the documents that are needed. What distinguishes it is that it separates three things that are usually mixed. What the documentation expressly says, what is reasonably deduced, and what does not appear anywhere.
iamericas-aclaraiaFree download, no sign-up
The skill starts asking about your system, not the model. You tell it what it does or what you want it to do, in your words, and you can attach a description or a PDF if you have one. If you attach something, read it before asking anything. Only when it understands your case does it show you the list of available models and you choose. It does not open documentation for models that you did not choose, it does not search for alternatives on its own and it does not make rankings. The report cites the document and the exact page where each item was found. When something does not appear, it says it as is, without completing what that supplier usually does or extrapolating from another version of the same model. Important: it is not a legal evaluation or a guarantee that the model will work well for your case. You may conclude that the documentation does not identify restrictions that prevent your use. You cannot conclude that the model will respond well to your customers.
Pick your platform and follow the steps. Where the tab is labelled “Native” the file installs as-is; ChatGPT and Gemini do not install it yet, so there you paste the content of SKILL.md into the chat.
It is installed. Turn it on from the capabilities panel, or type @ followed by its name inside any conversation.
C:\Users\YOUR-USERNAME\.claude\skills/Users/YOUR-USERNAME/.claude/skillsOn Windows, YOUR-USERNAME is the name of your account. Paste the path into the File Explorer address bar and it takes you there.
It is installed. Type /iamericas-aclaraia, or describe your case and Claude loads it on its own when it fits.
C:\Users\YOUR-USERNAME\.agents\skills/Users/YOUR-USERNAME/.agents/skillsOn Windows, YOUR-USERNAME is the name of your account. Paste the path into the File Explorer address bar and it takes you there.
It is installed. Restart Codex if it was open. Then type $iamericas-aclaraia, or /skills to list them all.
The scripts that compute scores do not run this way, so any numbers are the model's estimate. If you need the exact calculation, use it in Codex or Claude.
The scripts that compute scores do not run this way, so any numbers are the model's estimate. If you need the exact calculation, use it in Codex or Claude.
C:\Users\YOUR-USERNAME\.copilot\skills/Users/YOUR-USERNAME/.copilot/skillsOn Windows, YOUR-USERNAME is the name of your account. Paste the path into the File Explorer address bar and it takes you there.
It is installed. Type / in the chat and pick it from the menu. To share it with your team, put it in .github/skills/ inside the repository.
v1.2
21 September 2026 · 1.4 MB
Changes how the conversation starts. If the first message brings no context, the skill explains how the process works before asking anything: the phases, that it can be paused and resumed, and what you get at the end; and it notes that the input can be typed or attached as a file. If the context is already there, it confirms in two lines and gets going without that explanation. The description is rewritten to state that this is a guided, step-by-step process. Tests for both entry paths are added.
v1.1
14 September 2026 · 1.4 MB
Neutral Spanish throughout. The documentation of all 21 models moves from PDF to Markdown — the same information in 117 text files instead of 284 PDFs — so the download drops from 37.6 MB to 1.4 MB.
v1.0
1 August 2026 · 37.6 MB
First version.
It accompanies you to choose between several AI tools with evidence and leaves you a decision memo and the exact questions you have to ask.

Measurement cookies
We use analytics to learn which skills get downloaded and improve the catalogue. No advertising, no profiling.