Compare AI vendors on data use, exit, and support the user can point to in a document. Use when the user mentions AI vendor, model vendor review, which LLM vendor, AI procurement, or asks for a vendor note. AI in the business skill by Yasir Jilani.
Installs into .claude/skills of the current project.
Are you the author of Ai Vendor Note?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/syasj-ai-vendor-note)
---
name: ai-vendor-note
description: "Compare AI vendors on data use, exit, and support the user can point to in a document. Use when the user mentions AI vendor, model vendor review, which LLM vendor, AI procurement, or asks for a vendor note. AI in the business skill by Yasir Jilani."
license: MIT
compatibility: Agent Skills standard. No network access, extra packages, or credentials required.
metadata:
author: Yasir Jilani
version: "1.0.0"
domain: ai
---
<!-- GENERATED FILE - edits here are overwritten by scripts/generate.py.
Edit the 'ai-vendor-note' entry in source/, then run:
python3 scripts/generate.py && python3 scripts/validate.py
See CONTRIBUTING.md. -->
# AI Vendor Note
Compare AI vendors on data use, exit, and support the user can point to in a document.
## When to use this skill
Use this skill when the user:
- AI vendor
- model vendor review
- which LLM vendor
- AI procurement
## When not to use this skill
- The user wants a different domain's specialist skill.
- The task requires a licensed professional to decide, and the user only needs a referral note rather than a draft.
- The request asks you to deceive, evade a control, or hide material facts.
## Professional boundary
Do not write prompts or workflows that weaken safety rules, hide required disclosure, or invent model scores. This is not a certification and not a reason to send private data to a vendor.
## Operating boundaries
- Use only information the user provides or files they explicitly ask you to read. Do not invent metrics, laws, citations, prices, credentials, or clinical facts.
- Do not ask for passwords, API keys, tokens, seed phrases, one-time codes, or payment card data.
- Do not send data to an external service, install packages, or add network calls as part of this skill.
- Separate facts, assumptions, and recommendations. If a required input is missing, state the assumption or ask one focused question.
- If the user asks you to deceive a person, evade a control, forge a record, or cause harm, stop. Offer a legitimate alternative.
- Work product that affects money, employment, health, safety, or legal rights is a draft for a qualified human to review before it is used.
## Inputs to collect
- The vendors
- The documents they have
- Data-use questions
- How they would leave
## Workflow
### 1. Step 1
Compare only documents they uploaded.
### 2. Step 2
Mark a missing security or data term as missing.
### 3. Step 3
Ask how content is retained and whether it trains a model, and record their answer or the gap.
### 4. State the exit
can they take prompts and outputs with them.
### 5. Step 5
Do not invent a certification.
### 6. Step 6
Recommend a pilot limit, not a company-wide switch, when terms are thin.
## Output
Deliver a **vendor note**.
- Purpose of this vendor note, in two sentences.
- Facts the user supplied, listed separately from assumptions.
- The work itself, in the structure the workflow names.
- Open questions, risks, and the single next action with an owner.
- What a qualified reviewer still needs to confirm, if the domain is regulated.
## Quality bar
- Every number, date, name, and citation came from the user or is marked as an assumption.
- The artifact can be used without reading this skill again.
- Recommendations are specific enough that someone could accept or reject them.
- Boundaries were respected: no credentials requested, no unsupported professional claim, no deception.
## Example
### Scenario
Fieldnote is looking at two model vendors. One website shows a SOC 2 badge. Neither report is in the folder. Jonah has a one-page quote from each.
### Example data
```text
vendor A quote: $0.01 per call, 2 Sep 2026, data terms not attached
vendor B quote: $0.04 per call, 3 Sep 2026, says "no training" in the email, no contract
SOC 2 reports: none in the folder
exit question: unanswered by both
pilot limit Jonah wants: one queue, 30 days
```
### Example outcome
**Vendor note**
Do not treat the badge as a report. It is not in the file.
| Question | A | B |
| --- | --- | --- |
| Price in hand | $0.01 / call | $0.04 / call |
| Training use | not in the file | email says no, contract missing |
| SOC 2 | badge only | not mentioned |
| Exit with prompts | unanswered | unanswered |
Pilot, if any: one queue, 30 days, no card data, no company-wide switch.
Next: ask both for the data term in a document before a second pilot week.
## Anti-patterns
- Invented SOC reports
- A logo count as proof
- Ignoring data retention
## Related skills
- `vendor-security-review`
- `ai-data-note`