Use Guidelines

Manus claims "transform your thoughts into actions" It uses an Agent Loop Process. Users only need to input the topic or task description, and the system can automatically complete the process from content collection, structure generation, visual design to final output.

• Credit system

Manus has a limitation. It uses the "credit" system to restrict the number of tasks that users can generate within a period.

This is the usage flow chart generated by Manus

PlayBook

Manus also offers a collection of "playbook" templates to help users quickly conceive and implement ideas. This playbook covers a variety of areas, including Work & Life, Sales & Marketing, and Education, etc. helping users from different backgrounds improve the quality of their generated content.

                            For more specific functions and usage procedures, please visit the function page

<aside> 💡

Apart from the functions introduced in the Scenarios and Functions section, here are some more functions:

• The AI Profile Builder: It helps chemistry educators gather, synthesize, and structure detailed information on individuals, class members, organizations, concepts, or complex topics, such as specific chemical elements, scientists, advanced theories, or material science companies. Educators provide simple text input, documents, or URLs and it processes and organizes rich data into a comprehensive, detailed profile. This function supports research, class progress summaries, lesson planning, and curriculum development.
• The YouTube Influencer Finder: This tool helps chemistry educators efficiently discover high-quality educational content and reliable science communicators on YouTube. It identifies relevant chemistry channels and videos such as those focused on lab safety, chemical reactions, or sustainability while filtering out duplicates to save time. The system evaluates each channel’s focus, accuracy, and engagement metrics to ensure that recommended creators align with classroom goals and learning for students in targeted age. By providing organized results without subscription costs, this tool enables educators to integrate credible video resources into lessons, enhance student engagement, encourage the educational use of platforms like YouTube, and promote multimedia learning in chemistry education.

https://manus.im/playbook

</aside>

The Technology Behind Manus

The company does not officially provide the LLM model used by Manus. However, industry analysis indicates that Manus mainly relies on Anthropic's Claude Sonnet and Alibaba's Qwen models. (Sanchez, 2025)

However, the true strength of Manus lies in its agent-based orchestration system. Unlike other language models that focus on reasoning and text generation, Manus is an AI that employs a Multi-Agent Framework and operates as an autonomous agent. You can assign it a complex task in the morning, such as "creating a set of molecular orbital teaching slides with visual examples", and then come back later to find that it has completed. Therefore, the autonomous agent feature of Manus enables teachers to hand over the tedious teaching preparation (eg. content organization and visual design) to it, thereby allowing them to devote more time to teaching design and interaction with students.

The Manus development team mentioned in their official blog about their research on the “Context Engineering concept of Manus”(2025). This enables Manus not only to rely on the real-time conversation window, also through the mechanism of external memory and task layering to save more time understand the user's goals. This means it can read and update information between different tasks. For educators, Manus can "remember" the course goals and students' progress throughout the teaching cycle, helping teachers achieve continuous, project-based teaching support, rather than just being a one-time question-answering assistant.

Reference：

Manus AI. (2025, February ). Context Engineering for AI Agents: Lessons from Building Manus. Retrieved from https://manus.im/blog/Context-Engineering-for-AI-Agents-Lessons-from-Building-Manus

Sanchez, J. (2025, March 3). Manus AI Agent: What It Is, How It Works, & Its Impact. Retrieved from https://www.leanware.co/insights/manus-ai-agent