Skip to main content
Documentation

Learn Fleece AI

Learn how to set up, configure, and get the most out of your delegative AI workspace. Pick a topic below to dive in.

Getting Started

Go from zero to your first automated workflow in under five minutes.

1. Create your account

Sign up at fleeceai.app with your email, Google, GitHub, or Microsoft account. The 7-day free trial gives you 1,000 credits and 3 agents to get started.

2. Connect your apps

Open Settings > Integrations and connect the tools you use every day (Gmail, Slack, Notion, Google Sheets, etc.). All connections are OAuth-based and take seconds — no API keys to manage.

3. Start a chat or create a flow

Head to Chat to give your agent a one-off task, or open Flows to build a recurring automation. Try something simple first -- for example: "Every Monday at 9 AM, pull my open GitHub issues and post a summary to Slack."

Chat & AI Agents

Interact with frontier AI models and let agents execute tasks on your behalf.

Choosing a model

Fleece AI gives you access to frontier models. The Starter plan includes Mistral Medium 3.1 (French AI, GDPR-friendly). Pro unlocks GPT-5.2. Business adds GPT-5.4 and Claude Opus 4.7. Select your preferred model from the model picker at the top of any chat. Each model has different strengths — Mistral Medium 3.1 offers EU data residency, Claude excels at analysis and writing, GPT-5.4 offers the most advanced reasoning.

How agents work

When you send a message, the AI agent reads your instruction and decides which tools to call -- search the web, read a document, call an API, or run code. The agent loops until the task is complete and returns results in the chat.

Tool usage

Agents have access to all connected integrations as tools. They can send emails, create calendar events, update spreadsheets, post to Slack, and much more -- all from a single conversation. You can see each tool call in the chat timeline.

Flows (Automated Workflows)

Build recurring automations that run on a schedule or on demand.

Creating a flow

Open the Flows page and click "New Flow." Give it a name and describe what it should do in plain English. The AI agent interprets your description and builds the execution plan automatically.

Scheduling with cron

Every flow can have a cron schedule. Use presets like "Every day at 9 AM" or write a custom cron expression (e.g. 0 9 * * 1-5 for weekdays at 9 AM). Schedules are evaluated in UTC by default.

Execution modes

Flows support two execution modes. "Auto" lets the agent run without intervention -- great for trusted, repeatable tasks. "Always Ask" pauses before each execution and waits for your approval, giving you a human-in-the-loop safety net.

Monitoring runs

Each flow execution is logged with status (success, failed, pending), runtime duration, and full output. Open any run to inspect tool calls, intermediate results, and errors. Failed runs include retry options.

Integrations

Connect 3,000+ apps and manage your accounts.

How integrations work

Fleece AI uses the Model Context Protocol (MCP) to connect to thousands of third-party APIs. When you connect an app, Fleece AI handles OAuth tokens, refresh cycles, and API authentication so you never have to manage credentials manually.

Connecting an app

Go to Settings > Integrations > Connect New App. Search for the service you want (e.g. Slack, GitHub, Stripe), then follow the OAuth flow. Once connected, the integration is available as a tool in all your chats and flows.

Managing connected accounts

View all connected accounts in the Integrations tab. You can disconnect an app at any time, which revokes the OAuth token and removes tool access. If a token expires, Fleece AI will prompt you to re-authenticate.

Billing & Plans

Understand the Starter, Pro, Business, and Enterprise plans and what each includes.

Starter plan (€49/month)

The Starter plan includes a 7-day free trial with 1,000 credits, then 2,000 credits/month. You get 3 agents, 5 flows, 20+ essential integrations, Mistral Medium 3.1, custom skills & knowledge files, 30-day history, and email support. Save 20% with yearly billing at €39/month. Designed for individuals getting started with AI automation.

Pro plan (€99/month)

Pro unlocks 10 agents, 25 active flows, 8,000 credits per month, 1,000+ integrations, Mistral Medium 3.1 & GPT-5.2, heartbeat scheduling, inter-agent hierarchy, 60-day history, 3 team members, priority support (4h SLA), and API access. Save 20% with yearly billing at €79/month. Ideal for professionals and growing teams.

Business plan (€199/month)

Business includes 25 agents, 50 active flows, 20,000 credits per month, all 3,000+ integrations, Mistral Medium 3.1 + GPT-5.2 + GPT-5.4 + Claude Opus 4.7, 90-day history, 10 team members, audit logs, dedicated support (1h SLA), and Setup Express included. Save 20% with yearly billing at €159/month. Built for organizations with advanced needs.

Enterprise plan

Enterprise offers custom limits, SSO/SAML, SOC 2 Type II (in progress), a dedicated account manager, and on-premise deployment options. Contact our sales team for a custom quote.

Managing your subscription

Upgrade, downgrade, or cancel anytime from Settings > Billing. Plan changes take effect immediately. Billing is handled securely through Stripe -- Fleece AI never stores your card details.

API Reference

Access Fleece AI programmatically with API keys and REST endpoints. Available on Pro plan and above.

Overview

The Fleece AI REST API lets you manage agents, trigger flows, and retrieve execution results from any application. API access requires a Pro, Business, or Enterprise plan. Authenticate with a Bearer token generated from Settings > API Keys.

Full API documentation

For the complete endpoint reference with request/response schemas, code examples (cURL, JavaScript, Python), rate limits, and error codes, see the dedicated API Reference page at /docs/api. You can also download the OpenAPI 3.1 specification at /openapi.json to generate client libraries or import into Postman.

Security & Privacy

How Fleece AI protects your data, handles encryption, and stays compliant.

Data encryption

All data is encrypted in transit (TLS 1.3) and at rest (AES-256). Chat messages, flow configurations, and execution logs are stored in encrypted databases with strict access controls.

Third-party credentials

OAuth tokens for connected integrations are stored in encrypted infrastructure (SOC 2 Type II certification in progress). Fleece AI does not have direct access to your raw tokens — credentials are encrypted and isolated.

GDPR compliance

Fleece AI is GDPR-compliant. You can request a full data export or account deletion at any time. We do not sell personal data to third parties. For details, see our Privacy Policy.

Responsible AI

All AI interactions are subject to content safety filters. Flow executions are logged and auditable. The "Always Ask" execution mode ensures a human approves every action before it runs.

Still have questions?

Reach out to our team or explore the blog for in-depth guides and tutorials.