askdona.com

What It Is

AskDona is an organizational knowledge platform for enterprises and research institutions. Its core goal is to turn information scattered across procedures, manuals, reports, contracts, meeting minutes, and technical materials into a searchable, traceable knowledge base that can support business decisions. It is not just a general-purpose chatbot; rather, it is an enterprise knowledge-processing tool built around RAG Knowledge, Batch Assessment, Gen AI Chatbot, and Deep Research.

Core Capabilities

Its main focus is the combination of LLMs and RAG. After users register files and URLs, the system breaks down, organizes, and structures the data, then generates answers with sources. The official site emphasizes “Citations, always,” meaning each answer can link back to source documents or be traced via citation numbers. dona-rag-2.0 is described as being able to handle long, complex professional documents while preserving structures such as conditions, exceptions, and notes. Typical use cases include internal knowledge Q&A, cross-document verification, review and confirmation workflows, research investigations, and embedding a chatbot based on a proprietary knowledge base into an internal portal or public website.

Pricing and Integrations

The current pages do not disclose specific pricing, plans, free quotas, or trial limits. They only provide entry points for booking a demo, requesting pricing information, and applying for a trial, which clearly suggests an enterprise-sales model. On the integration side, the clearest public information concerns the website chatbot: it can be embedded with a single script tag, without an SDK, build process, or self-hosting, and supports configuration of brand colors and display style. Public sites support CORS/Origin restrictions and reCAPTCHA v3. Beyond this, there is no visible information about an open API, webhooks, or connectors for common enterprise systems.

Pros, Cons, and Limitations

Its strengths are a clear product positioning and a strong emphasis on verifiability and source tracing. Its in-house RAG architecture highlights end-to-end control from document processing to evidence extraction, and its experience with RIKEN R-CCS and the Fugaku support site provides credibility for professional use cases. The limitations are also apparent: the underlying large language model is not disclosed, pricing and SLA details are opaque, and there is insufficient information on privacy compliance, whether data is used for training, and deployment options. Chinese-language support is not mentioned, and the site is mainly available in Japanese and English. While output quality is heavily emphasized, it still depends on the quality of the knowledge base, permission settings, and document coverage.

Who It’s For and Access from China

AskDona is better suited to enterprises, research institutions, and knowledge-intensive departments in the Japanese market, especially organizations that need to reduce repeated questions to experts, improve review consistency, and make internal documents reliably citable. Access from China cannot be determined from the main content, and payment methods are not disclosed. Chinese teams looking for alternatives may evaluate Dify, FastGPT, AnythingLLM, or cloud-vendor solutions such as Azure AI Search, Amazon Q Business, and Vertex AI Search.