Prompt Management

Prompt Management lets administrators view and customize the AI prompts and policies that control how the GrndCtrl chatbot responds for your organization. You can review the base system prompt (read-only), edit tenant-specific policies, and monitor token usage. Edits reach new conversations within about five minutes.

Administrator access required

You need to be a tenant administrator or a developer administrator to open Prompt Management.

Accessing Prompt Management

Click the Apps button — the grid icon in the header, available on every page — and choose Prompt Management. See Finding and Opening Your Apps if it is not listed there.

The page loads with three collapsible sections: Prompt Statistics, System Prompt, and Tenant Policies.

Prompt Statistics

Click the Prompt Statistics section header to expand it. This panel shows token and character usage at a glance:

Metric	Description	Color Coding
Estimated Tokens	Approximate token count for the full prompt	Green (<50k), Amber (<75k), Red (≥75k)
Total Characters	Combined character count across system prompt and policies	-
System Prompt	Character count of the base system prompt	-
Tenant Policies	Character count of all tenant policy text	-

Below the cards, a Policy Breakdown shows the character count for each of the three policies: Domain Context, Response Style, and Quality Assurance.

System Prompt (Read-Only)

The System Prompt section shows the base prompt that drives the AI. This prompt is managed by NXT and cannot be edited by tenant admins.

• The header shows the prompt version and the number of sections
• Each section is listed with its name and content
• The {client_policies} placeholder in the system prompt is where your tenant policies get injected

Why is this read-only?

The system prompt contains the core instructions that ensure the AI behaves correctly across all tenants. Your customizations happen in the Tenant Policies section below, which gets injected into the system prompt at the {client_policies} placeholder.

Tenant Policies (Editable)

This is where you customize how the AI responds for your organization. There are three policy sections, each containing editable text elements:

Each policy holds a fixed set of named elements — eleven in total. Every element is a block of free text with its own character limit, and the limit is shown as you type.

Domain Context

Who your organization is, and how the assistant should read your questions.

Element	What it sets
Tenant Identity	Your organization’s identity, expertise, and core services — this shapes how the assistant introduces itself.
Document Library	Your repository structure and key folders, so the assistant can navigate your knowledge base.
Query Interpretation	How to read ambiguous or industry-specific questions — common patterns and your interpretation rules.
Source Requirements	When citations are required and what format they take.
Professional Boundaries	When the assistant should recommend consulting a licensed professional instead of answering.

Response Style

Element	What it sets
Tone & Language	Communication style, formality, and the domain terminology to use naturally.
Response Structure	How different kinds of answer are organized — diagnostics, technical questions, general enquiries.
Brand Voice	Your preferred phrases and communication personality, with examples of on-brand language.

Quality Assurance

Element	What it sets
Ethical Standards	What must be true before the assistant delivers sensitive recommendations.
High-Stakes Topics	Subjects that require elevated validation — safety-critical, legally sensitive, or liability-affecting. This is the list to edit if you want the assistant to be more careful about a particular subject.
Disclaimer Template	The standard disclaimer for safety-critical recommendations, and when it is applied.

There is no allow-list or block-list of topics

Nothing here switches subjects on or off. The closest thing is High-Stakes Topics, which raises the standard of evidence for the subjects you name rather than refusing them, and Professional Boundaries, which says when to hand off to a human expert.

Editing a Policy

1. Expand the Tenant Policies section.
2. Click on the policy element you want to edit (e.g., Tenant Identity).
3. Modify the text in the textarea — a character counter shows {current}/{max}.
4. Click Save Changes to apply (the button appears when changes are detected).
5. Changes reach new conversations within about five minutes — policies are cached per organization, so an edit is not instant. Existing conversations are unaffected either way.

Reset cannot be undone, and it takes the whole policy

To restore the default text, click Reset to Default. You are asked to confirm — “Reset to default? This cannot be undone.” — and must click Confirm Reset. Reset applies to the whole policy you are looking at, not to the one element you were editing. Copy any wording you want to keep before you reset.

Troubleshooting

I can’t find Prompt Management

Solutions:

• You need tenant administrator or developer administrator permissions. Contact your administrator.
• Your Home screen may not have a card for it. Use the Apps button — the grid icon in the header — which lists everything you can reach.

Token count is in the red zone

Solutions:

• Your combined policies are very large (≥75k tokens). This can slow down AI responses and increase costs.
• Review each policy section and trim unnecessary text — focus on the most important guidelines
• Remove duplicated instructions across policies

My policy changes don’t seem to affect responses

Solutions:

• Policy changes apply to new conversations only — they don’t retroactively change existing chat threads
• Start a new chat to test your updated policies
• Make sure you clicked Save Changes — if the button is still visible, the change hasn’t been saved yet

Related Articles

• Documents and AI Chat
• Branding and Administration
• Getting Started with GrndCtrl