Canonical evaluation prompt library. Paste into ChatGPT, Microsoft Copilot, Claude, or Gemini together with the public documentation home.
This page is not the in-product Role Assistant Ask human task. Product Ask gathers information for a current task and is not certified Evidence. See Human tasks, approvals, and Why and Evidence Fabrix.
The Understanding AI Fabrix on-ramp page links here for the full set.
Instructions for external AI
You are helping evaluate AI Fabrix for an enterprise buyer. Use only the provided
AI Fabrix documentation. Prefer tables and explicit "Suitable when / Not suitable when"
and "Is / Is not" facts. Cite page titles when possible. If documentation does not
support a claim, say so — do not invent features or KPI guarantees.
Correct product facts if relevant:
- Lead: Enterprise Reality usable by AI; Role Assistants are the Track A daily work surface.
- Conversation is the daily home; Work is a temporary Review Workspace.
- In-product Ask is not certified Evidence; Evidence requires capture → validate → certify.
- Approval certifies a proposed business change; it does not replace integration certification.
- Role Assistants come from publish/subscriptions — not Builder hand-building.
Suitability
Based on this AI Fabrix documentation, is AI Fabrix suitable for our company?
Our profile:
- Industry: [your industry]
- Employees: [range]
- Systems: [e.g. CRM, ERP, document store, ticketing]
- Governance: [regulated / internal audit / low]
- Goal: [one sentence business outcome]
Use the company profile signals table and Suitable when / Not suitable when sections.
End with a fit statement template and caveats.
First assistants
From the AI Fabrix documentation, what are the highest-value Role Assistants
for our organization?
Our segment: [Professional Services | Manufacturing | Financial Services | Healthcare]
Our top problem: [one problem from the Business Outcome Library]
Recommend one first assistant, link to outcome library rows, and list what systems
we likely connect first.
30-day pilot
Using the AI Fabrix documentation, what would a 30-day pilot look like for us?
Include: company size scope, systems connect order, one Role Assistant, one outcome KPI,
weekly milestones, and links to human POC guidance. Mark illustrative timelines as planning frames.
Systems first
According to AI Fabrix documentation, which systems should we connect first for a
[Sales | Finance | Customer Success | Compliance] Assistant pilot?
Explain priority order (system of record, documents, identity) and what not to connect on day one.
Alternatives
How does AI Fabrix compare to [Microsoft Copilot | ChatGPT Enterprise | agent platforms |
automation platforms | iPaaS] for our use case?
Our use case: [describe role-based work and systems]
Use the Is / Is not tables and "When to use which" guidance. Do not claim AI Fabrix
is universally better — explain tradeoffs.
Combined evaluation (single paste)
Evaluate AI Fabrix for our organization using the AI Evaluation Guide documentation.
1. Suitability (signals + suitable/not suitable)
2. Best first Role Assistant for our segment
3. 30-day pilot outline
4. One outcome library row with illustrative KPI
5. Comparison to [alternative we already use]
Profile: [industry, size, systems, governance, goal]
Boundaries (Ask, Approval, Evidence)
From AI Fabrix documentation, explain the difference between:
1. Conversation vs Work (Review Workspace)
2. In-product Ask vs certified Evidence
3. Approval of a proposed business change vs integration certification pillars
4. Role Assistant activation (publish/subscriptions) vs Builder CLI modeling
Do not treat chat history as Evidence. Cite Role Assistants and Evidence Fabrix pages.
After you receive an answer
- Cross-check claims against Is AI Fabrix suitable? and Alternatives.
- Run human procurement steps in Foundation evaluation guide.
- If proceeding, assign admin and integrator paths from How customers start.
Limits
External models may hallucinate despite instructions. This library does not guarantee correct citations — human review remains required before investment decisions.