Project settings
Configure every aspect of your project from the edit page.
General
The General section contains the core fields that describe your project. These fields are visible to reviewers and inform the AI when generating improvements.
| Field | Description |
|---|---|
| Title | The name of your project, displayed everywhere. |
| Summary | A short description of what your project does. |
| Problem | The problem your project solves. |
| Solution | How your project addresses the problem. |
| Differentiation | What sets your project apart from alternatives. |
| Risks | Known risks or challenges the project faces. |
| Target users | Who the project is built for. |
| Product URL | A link to the live product or landing page. |
| Custom questions | A JSON field for defining custom reviewer questions shown in the feedback form. |
| Thumbnail URL | An image URL used as the project thumbnail in listings and embeds. |
Tip: The more detail you provide in the problem, solution, and differentiation fields, the better AI-generated improvements will be. These fields give the AI context about what matters most for your product.
Protection level
Control what reviewers must agree to before accessing your project. The protection level determines the gating applied to new invites.
| Level | Description |
|---|---|
| None | No protection. Reviewers can access the project immediately after accepting an invite. |
| Pledge | Reviewers must agree to a confidentiality pledge before accessing the project. |
| NDA Template | Reviewers must sign a standard NDA generated by Lupr before accessing the project. |
| Custom NDA | Reviewers must sign your own NDA text or uploaded document before accessing the project. |
Changing the protection level only affects new invites. Existing accepted invites keep their original gating — a reviewer who already accepted under a previous level will not be asked to re-sign.
NDA modes
When using the NDA Template level, you can choose between two modes:
One-way NDA
The reviewer agrees not to disclose information shared by the project owner. Suitable for most feedback scenarios.
Mutual NDA
Both parties agree not to disclose each other’s confidential information. Useful when reviewers may also share sensitive context.
Custom NDA
With the Custom NDA level, you can paste your own NDA text directly into the editor or upload a file. Reviewers will be presented with your custom agreement and must sign it before they can access the project.
Assumptions
The assumptions section is a hypothesis tracker for structured validation. Use it to define the core assumptions behind your product and track whether feedback actually validates them.
Fields for each assumption
| Field | Description |
|---|---|
| Hypothesis | A clear statement of what you believe to be true about your product or market. |
| Success criteria | The measurable conditions that would confirm the hypothesis. |
| Status | Tracks progress: Untested → Testing → Validated or Invalidated. |
| Evidence | Notes and references from feedback that support or refute the hypothesis. |
Status progression
The hypothesis has been defined but no evidence has been collected yet.
You are actively collecting evidence or running experiments to evaluate this hypothesis.
Evidence supports the hypothesis. The assumption holds true.
Evidence contradicts the hypothesis. Consider pivoting or revising your approach.
Tip: Assumptions are most useful when you define them before collecting feedback. As reviews come in, update the status and evidence to see whether your core beliefs hold up against real user input.
Domain
Customize how people access your project with a readable slug or your own domain.
Custom slug
Replace the auto-generated project ID with a readable URL. For example, instead of a random ID your project could be accessible at lupr.dev/p/my-app.
Custom domain
Use your own domain to host the project feedback page. For example, feedback.yourapp.com.
DNS setup
To configure a custom domain, add two DNS records:
CNAME record
Point your subdomain to Lupr’s servers for routing.
TXT verification record
Prove ownership of the domain so Lupr can issue an SSL certificate.
For more details on API access with custom domains, see the API documentation.
Badge
The badge is a public-facing indicator you can embed on your site. It displays your validation score, feedback count, and validation status.
Enabling the badge
Toggle the badge on or off from the project settings. When enabled, the badge unlocks public API access including the embed widget, REST API endpoints, SVG badge, and public feedback page.
Important: The badge must be enabled for the embed widget, REST API, and badge embed to work. If the badge is disabled, all public-facing endpoints return no data.
What the badge shows
Validation score
Your validation score calculated from feedback responses.
Feedback count
The total number of feedback submissions your project has received.
Validation status
Whether your project has reached validation thresholds based on assumptions.
Widget
The widget is an embeddable component you can add to your site to display project feedback, scores, and validation status. Configure what data is shown from the project settings.
For embed instructions and code snippets, see the API documentation.
Integrations
Connect external services to your project to dispatch feedback and improvements to your existing tools.
GitHub
Connect a repository. Shows connection status and allows dispatching issues.
Linear
Connect a workspace. Dispatch feedback and improvements as Linear issues.
Slack
Connect for notifications when feedback is submitted or improvements are generated.
For detailed setup instructions, see the Integrations documentation.
Dispatch rules
Dispatch rules automate the process of sending feedback and improvements to external tools. Define triggers, conditions, and targets to have items dispatched automatically when your criteria are met.
For a full breakdown of triggers, conditions, targets, and rule history, see the dispatch rules section in the Integrations documentation.
Policy overlay
The policy overlay lets you customize AI agent behavior for this specific project. By default, agents follow your account's tier policies, but you can override those defaults at the project level to fine-tune how the core loop processes feedback and generates improvements.
For details on how policies work and the core loop pipeline, see the Core Loop documentation.
Danger zone
The danger zone contains destructive actions that cannot be reversed.
Delete project
Deleting a project permanently removes all associated data, including feedback, improvements, intents, and agent tasks. This action is immediate and cannot be undone.
Warning: Deleting a project is permanent and cannot be undone. All feedback, improvements, and agent tasks will be deleted.