App Guidelines

Quality Standards

Every plugin on the CableKnit marketplace and runtime should solve a real business problem. We expect:

• A clear, accurate description of what the plugin does
• At least one working automation template
• Documentation that explains setup and usage
• A pricing tier that reflects the value delivered

What We Look For

• Proper automation workflows with clear triggers and actions
• Documentation pages covering setup, configuration, and expected behavior
• Reasonable pricing that meets the platform cost floor
• A clear value proposition — why would a business install this?
• Correct use of platform tools and connectors

What Gets Rejected

• Empty or stub plugins with no real functionality
• Misleading descriptions that don't match what the plugin actually does
• Missing documentation
• Prompt injection or attempts to manipulate the AI runtime
• Duplicates of existing marketplace plugins with no differentiation
• Low-effort submissions that don't meet basic quality standards
• Pricing below the platform cost floor

Pricing Rules

Every plugin must meet the platform cost floor, which is calculated based on the AI resources your automations consume. The cost floor is shown in your bundle validation output.

Beyond the cost floor, pricing should reflect the value your plugin delivers. A plugin that saves a business 20 hours per week is worth more than one that saves 2.

Submission Process

1. Build and validate your bundle with cableknit validate
2. Push your bundle with cableknit push
3. Submit for review from the developer portal
4. We review within 1–3 business days
5. You'll receive an email with approval or feedback for revision

After Approval

Approved plugins go live on the marketplace and are available for businesses to install. If you re-push a new bundle via the CLI, the plugin resets to draft status and requires a new review before going live again.

Keep your plugin updated and responsive to user feedback. We may reach out if we notice issues post-approval.