No-code Conversational AI Building Service "miibo" Releases New "Thinking Mode" Feature, Enabling AI Agents to "Autonomously and Repeatedly Deliberate"

miibo Inc. (Head office: Minato-ku, Tokyo; Representative Director: Masashi Kunugi) announced the release of "Thinking Mode," a new feature in its no-code conversational AI service "miibo," which enables AI agents to "autonomously and repeatedly deliberate."

What is Thinking Mode?

Traditional AI chatbots typically provide an immediate, single response to a user's question. However, this method may not be sufficient for complex questions or tasks requiring multi-stage processing.

With miibo's newly implemented Thinking Mode, AI agents autonomously make judgments internally and process information step-by-step, as shown in the example below.

User: "Help me prepare for next week's business trip to Tokyo."

◼︎AI Agent's Thought Process:

├─ Step 1: Confirm trip dates

├─ Step 2: List necessary reservations (transportation, accommodation)

├─ Step 3: Create a packing checklist

├─ Step 4: Check weather forecast and advise on clothing

└─ Complete: Organize all information and respond

AI Agent: "Regarding your business trip to Tokyo next week, I have summarized the following preparations..."

Features of Thinking Mode

1. Planned Task Execution

In Thinking Mode, the AI agent first creates a to-do list and processes each item reliably. This ensures that complex requests are handled without omissions.

2. High-Quality Answers through Deep Thinking

Instead of simple one-shot Q&A, the agent considers multiple perspectives, generating more accurate and comprehensive answers.

3. Transparent Thought Process

The log function in the administration screen visualizes how the AI agent thought and what steps it took. This allows users to review the AI's decision-making process and use it for improvements.

4. Deeper Answers through Integration with Knowledge, MCP, and Connectors

Thinking Mode truly shines when combined with miibo's other features: "Knowledge Data Store (RAG)," "MCP (Model Context Protocol)," and "Connector Function."

A. Deep Answers Across Multiple Information Sources

Traditional RAG generated answers using only information retrieved in a single search. In Thinking Mode, as shown in the example below, the agent searches knowledge repeatedly as needed, building up information to construct an answer.

User: "Create a competitor comparison report for new product A."

◼︎AI Agent's Thought Process:

├─ Step 1: [Knowledge Search] Retrieve specifications and features of new product A

├─ Step 2: [Knowledge Search] Retrieve information on competitor products B, C, D

├─ Step 3: [Knowledge Search] Confirm format of past comparison documents

├─ Step 4: Summarize collected information into a comparison table

├─ Step 5: Organize the advantages of our company's product

└─ Complete: Generate competitor comparison report

AI Agent: "I have created the competitor comparison document. New product A is..."

B. Autonomous Task Execution Integrated with External Systems

By combining with MCP and Connectors, the agent can automate a series of processes from information retrieval to processing and action execution, as shown in the example below.

User: "Aggregate this month's sales and share the report on Slack."

◼︎AI Agent's Thought Process:

├─ Step 1: [Connector] Retrieve sales data from the sales management system

├─ Step 2: Analyze data, calculate month-over-month change and trends

├─ Step 3: [Knowledge Search] Confirm format of past reports

├─ Step 4: Create a report including graphs and a summary

├─ Step 5: [MCP] Post the report to the specified channel via Slack API

└─ Complete: "Sales report created and shared in #sales channel."

How to Switch to Thinking Mode

Thinking Mode can be easily configured from the sidebar of the administration screen.

The setup procedure consists of just 2 steps:

1. Open "Models/Prompts" in the sidebar and check "Use Thinking Mode" within "Thinking Mode Settings."

2. Set instructions related to agent behavior, such as tool usage and multi-step processing, in the "Thinking Mode Prompt."

That's it for the setup. No coding or specialized knowledge is required. Anyone can immediately start building AI agents using Thinking Mode with no code.

Use Cases

It can be utilized more conveniently in the following scenarios:

1. Complex Research Requests

Example: Automated generation of market research/competitor analysis reports For requests such as "Summarize competitor pricing strategies in the domestic SaaS market," the system repeatedly searches and integrates competitor information, past research reports, and industry trend data from the Knowledge Data Store, automatically generating a comprehensive analysis report with a single instruction. While traditional RAG generated answers using only the results of a single search, Thinking Mode provides highly accurate research results by accumulating information.

2. Multi-stage Business Support

Example: Automation of recruitment tasks/internal application flows

With instructions like "Proceed with applicant A's document screening and send the results to Slack," the AI agent autonomously plans and executes a series of steps: ① confirming applicant information, ② matching against selection criteria, ③ assisting with pass/fail judgment, ④ generating result text, and ⑤ sending to Slack. This eliminates the need for staff to give instructions at each step, allowing complex business flows to be fully entrusted to the AI.

3. Specialized Consultations

Example: Internal specialized consultation desk for legal/accounting, etc.

For highly specialized questions such as "Please confirm if this contract clause is consistent with our company's regulations," the system cross-references internal regulations, past contract examples, and relevant legal information from multiple knowledge sources, providing specific advice beyond generalities. This reduces inquiry workload for relevant departments while providing evidence-based answers instantly.

4. System Integration Tasks

Example: End-to-end automation from sales report aggregation to Slack sharing

For instructions like "Aggregate this month's lead acquisition numbers from CRM and report them to the sales channel along with month-over-month comparison," the system automatically processes a series of steps: ① data retrieval from CRM system (Connector), ② aggregation and month-over-month calculation, ③ referencing past report formats (Knowledge Search), ④ report generation, and ⑤ posting to Slack (MCP). This allows routine tasks spanning multiple systems to be completed with a single instruction.

Summary

Thinking Mode is a feature that enables AI agents to autonomously and repeatedly exercise their "thinking power."

・ Autonomously plans and executes complex tasks

・ Searches knowledge multiple times to generate deep and accurate answers

・ Operates external systems in conjunction with MCP and Connectors

・ Visualizes transparent thought processes

AI agents autonomously think and judge to lead to the optimal answer—that's Thinking Mode. Expand the possibilities of AI agents even further with miibo's Thinking Mode.

About miibo Inc.

miibo Inc. develops and provides the conversational AI building platform "miibo" with the mission of "enriching people's lives with AI technology." As shown in this case, we are working to solve social issues by utilizing AI technology.

https://miibo.co.jp

About the no-code conversational AI building service "miibo"

The no-code conversational AI building service "miibo" by miibo Inc. is a service that allows anyone to easily create practical conversational AI without code.

Official website: https://miibo.ai

・ Easy AI Application Creation for Anyone

・ No difficult skills or languages required! It's a conversational AI building service that allows you to quickly create AI-powered applications using your existing databases and Large Language Models (LLMs).

・ Super Agile Development: Create and Test

・ Rapid implementation, proof-of-concept, effectiveness verification, and refinement. Accelerate the entire PDCA cycle of development repeatedly. Enables speedy proposals to customers.

Conversational AIs applicable to various uses are being created daily using miibo, and its adoption is progressing in listed companies, government agencies, and local governments.