MindPal

What it does

MindPal takes in documents, SOPs, frameworks, training materials, brand guidelines and web URLs uploaded to a workspace. From that knowledge it builds AI agents that answer in your voice, and workflows that chain agents and other steps into repeatable processes. Published tools can be shared as chatbots, embedded tools or branded portals.

Most people start with one document and one task, then test the agent against realistic scenarios before sharing it. Guardrails, human checkpoints for high-stakes decisions, and per-tool sharing controls help manage it. Workflows support branching, loops and human input between steps.

MindPal suits coaches, consultants, course creators and agencies who want to offer a branded AI experience to clients. The free plan includes 100 AI credits and does not require a credit card. Paid plans add credits, storage and seats, and usage depends on the model and steps executed. Model output can still be wrong, so review and human oversight remain necessary.

What it can do

• Build AI agents grounded in uploaded documents, SOPs and frameworks
• Chain agents and steps into multi-agent workflows with branching, loops and human input
• Set guardrails and review conversations
• Publish agents as links, embeds, or branded portals
• Use custom branding on paid plans; custom domains are an add-on on Pro and Advanced
• Bring your own model API key (BYOK) with an active paid plan

Example workflows

Client onboarding agent

Input: onboarding walkthrough, SOP documents and brand guidelines uploaded to a workspace. Steps: build an agent on that knowledge, set guardrails and test it with representative client questions, then publish it as a link or embed. Output: a branded chatbot clients use to get guided onboarding answers.

Multi-step deliverable workflow

Input: a discovery call script, case study template and coaching framework. Steps: create one agent per stage, connect them in a workflow with inputs and outputs, add a human input step for review, and test the full workflow. Output: a repeatable process producing a structured client deliverable, with a human checkpoint before release.

Before you choose

• Model output can be incorrect even when grounded in your knowledge; review outputs and add human review for consequential decisions.
• Review data-security and model-provider policies before uploading sensitive information. AI credit costs depend on the model and steps executed.
• Free and paid plans differ in credits, storage, seats and custom domains; custom domains are an add-on on Pro and Advanced.

Listed under

• Coding
• MCP and agent tools

Get started

Free plan with 100 AI credits.

1. Create a free workspace and upload one document such as a framework or SOP
2. Build an agent with instructions, brand voice and guardrails, then test it against real scenarios
3. Publish the agent through a link or embed, configuring sharing settings and seats as needed

Free includes 100 AI credits; Stripe workspaces receive a monthly allowance, legacy Lemon Squeezy Free workspaces a one-time starter allowance. Add-ons: extra AI credits $9 per 1,000 one-time; extra storage $1/GB monthly; custom domain $9 monthly; extra Editor seat $29 monthly; extra User seat $5 monthly. BYOK requires an active paid plan and model-provider charges are separate. Public API from Pro on standard paid plans. Annual plans billed for the full year.

Source: mindpal.space/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Turn your expertise into 24/7 AI agents and multi-agent workflows | MindPal ↗mindpal.space
• MindPal Pricing — Free Plan Available | Start Building AI Agents Today ↗mindpal.space

Company website

mindpal.space: product, pricing and policy pages

YC directory mirror

Batch and company name

Human review

Checked by a person

knowledge_base

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.