7. Frequently Asked Questions (FAQ)
If you are exploring or already using Treeify, this guide will help you quickly understand its core capabilities, usage flow, file parsing support, generation logic, integration options, pricing model, and feedback channels.
1. Product Overview and Basic Usage
Q1. What is Treeify?
Treeify is an AI-powered test design and test case generation tool built to help teams complete test design work with higher efficiency and more consistent quality.
Unlike one-shot generation tools, Treeify emphasizes a structured, reviewable, editable, and continuously improvable test design process. Generated results are presented in an editable mind map, making it easier for QA teams to inspect structure, review coverage, and refine outputs iteratively.
Q2. Can I use Treeify without uploading files?
Yes.
In addition to file upload, you can directly enter requirement information in the text input area provided in the product. Treeify can then generate test objects and test scenarios based on your input and tailor the output to your description.
This means that even if you do not yet have a prepared Word document, PDF, or spreadsheet, you can still start by entering the core requirements manually.
Q3. Does Treeify offer a free trial?
Yes.
During the current beta stage, each account receives 100 free credits to explore Treeify’s core capabilities.
2. Generation, Editing, and Continuous Optimization
Q4. What should I do if the generated results are not detailed enough or not extensive enough?
You can ask Treeify to generate more.
First, select the part you want to expand, then enter a more specific request in the chat box, such as:
- Please generate more fine-grained content
- Please generate more requirements
- Please generate more test objects
- Please generate more test cases
Treeify will further refine or expand the results based on the selected scope and your instructions.
Q5. Editing JSON or tables directly is not very convenient. Is there a better way to modify the results?
Yes. We recommend using the chat-based editing workflow.
You can first select the content you want to revise, then describe your requested changes in natural language. Treeify will return updated results based on your instructions, and you can review them before replacing the original content.
Compared with directly editing JSON or tables, this approach is more efficient and better suited to complex logic changes.
Q6. Can I edit the generated test cases?
Yes.
You can review and edit any generated content through the mind map editor, the detailed editing form provided in the product, or the chat-based editing workflow.
Treeify is not designed to simply generate results once and leave you with them. Its goal is to help you review, refine, and iterate on structured outputs more efficiently.
Q7. I want Treeify to learn the logic behind my revisions so I do not have to repeat them next time. How can I do that?
You can.
First, select the content you want to revise or optimize, then complete the adjustment through chat. After that, click Generate Skills. Treeify will summarize the logic behind your changes and generate the corresponding skills automatically.
In similar future scenarios, Treeify will evaluate when to apply those skills, reducing repeated manual adjustments and gradually aligning with your test design standards and habits.
Q8. Can I upload my own skills?
This capability is currently under development.
Once it is available, we will announce it in the beta user group.
3. Evidence Semantics and Missing Information
Q9. What is Evidence Level (evidence_level)?
Evidence Level (evidence_level) indicates how strongly a generated result is supported by the input.
It helps you understand whether a piece of content is:
- explicitly stated in the input
- inferred from the surrounding context
- classified by AI into a certain type or structure
- supplemented based on common domain knowledge
This makes it easier to distinguish between content that can be used more directly and content that may require further review.
Q10. What types are included in Evidence Level (evidence_level)?
Evidence Level (evidence_level) mainly includes the following four types:
- Explicit Evidence (
explicit) - Implied Evidence (
implied) - Type Inference (
inferred_type) - Domain-Common Evidence (
domain_common)
These categories represent different evidence sources and confidence levels, helping you understand where a generated result comes from and how it should be interpreted.
Q11. What does Explicit Evidence (explicit) mean?
Explicit Evidence (explicit) means that the content is directly and clearly stated in the input material.
For example, the information already appears in the requirement document, specification, business rule, or prototype, and Treeify is simply extracting, organizing, or structuring it without adding further inference.
This is the highest-confidence evidence type.
Q12. What does Implied Evidence (implied) mean?
Implied Evidence (implied) means that the content is not written word for word in the input, but can be reasonably derived from the available context.
It is not invented content. Instead, it is inferred from process logic, field semantics, contextual relationships, or state transitions.
This type is still grounded in the input, but compared with Explicit Evidence (explicit), it is more suitable for confirmation in critical cases.
Q13. What does Type Inference (inferred_type) mean?
Type Inference (inferred_type) means that the underlying fact may come from the input, but the test type, analytical dimension, or structural classification of that content is inferred by Treeify.
For example, the input may explicitly state a field length limit, while the judgment that this belongs to boundary condition testing may be classified as Type Inference (inferred_type).
This is not fabricated business logic. It is Treeify’s structural interpretation of the test design.
Q14. What does Domain-Common Evidence (domain_common) mean?
Domain-Common Evidence (domain_common) means that the content is not directly written in the input and not strictly derived from input logic, but is added based on common domain rules, industry practices, or typical test concerns.
For example, a requirement may describe only the normal flow, without mentioning invalid input, empty values, illegal characters, or common security restrictions. Treeify may supplement these areas based on standard testing practice.
This type of content often improves coverage, but it does not mean that the project has explicitly required it.
Q15. Why are additional fields required when Evidence Level (evidence_level) is Type Inference (inferred_type) or Domain-Common Evidence (domain_common)?
Because these two categories are not facts that are explicitly committed in the input.
They represent Treeify making an additional design judgment or supplementing content based on broader test knowledge. To keep the result transparent, reviewable, and correctable, Treeify requires two extra fields:
- Why It Is Needed (
why_it_is_needed) - How To Correct (
how_to_correct)
Specifically:
- Why It Is Needed (
why_it_is_needed) explains why this addition or classification is necessary - How To Correct (
how_to_correct) explains how to revise it if it does not match the real project situation
This helps you understand why the content appears and how to adjust it when needed.
Q16. What is Missing Text Placeholder (MISS_TEXT_FIELD)?
Missing Text Placeholder (MISS_TEXT_FIELD) is used to indicate that a required and important piece of information is missing from the input.
When Treeify determines that a field or a testing conclusion depends on important information that is not present in the input, it does not guess and does not hide the gap with vague wording. Instead, it outputs MISS_TEXT_FIELD directly.
This placeholder must be preserved exactly as is. It must not be translated, rewritten, or replaced.
Q17. Why does the generated result contain Missing Text Placeholder (MISS_TEXT_FIELD)?
Because one of Treeify’s core principles is this:
It is better to expose a real information gap than to fill it with hallucinated content.
Many generative systems will produce a plausible-looking answer when information is missing. However, this can make users believe the content is grounded in actual requirements when it is not. Treeify takes the opposite approach: when critical support is missing, it marks the gap explicitly with Missing Text Placeholder (MISS_TEXT_FIELD).
This usually happens when:
- a test conclusion depends on key information that is missing
- the input mentions something relevant but does not define it clearly enough
- the system cannot infer the content safely, even if a guess seems possible
The purpose is not to say that “the model failed to finish.” The purpose is to help you identify where requirements need to be supplemented, clarified, or corrected.
4. File Parsing and Data Security
Q18. Can Treeify recognize images inside uploaded files?
Yes.
Treeify supports image recognition and understanding for content embedded in files, including flowcharts, interface screenshots, structure diagrams, and similar materials. In other words, Treeify can perform multimodal analysis by combining text and image information.
Please note that image recognition usually consumes more tokens, so the parsing cost is generally higher than for text-only files.
Q19. What file types does Treeify currently support?
Treeify currently supports the following file types:
- PDF (
pdf) - XMind (
xmind) - Markdown (
md) - Text (
txt) - CSV (
csv) - Excel (
xls,xlsx) - Visio (
vsd,vsdx) - Word (
doc,docx)
The text content in these files can be used in Treeify’s downstream analysis and generation workflow. For some file types, embedded images can also be interpreted through multimodal parsing.
Q20. What kinds of files can I upload?
You can upload files containing business requirements, system design information, user stories, or other content relevant to test design, such as Word documents, PDFs, and spreadsheets with requirement details.
As long as the file provides useful input for generating test objects and test scenarios, Treeify can parse and process it.
Q21. Is my uploaded data secure?
Yes.
All uploaded files are encrypted during transmission and stored securely. Treeify does not retain your data beyond the time required for processing.
If your organization has higher security requirements, you can also explore Treeify’s private deployment option.
5. Export, Integration, and Deployment
Q22. What export formats are supported?
Treeify currently supports exporting results in the following formats:
- Excel
- JSON
Q23. Does Treeify support MCP integration with existing enterprise workflows?
Yes.
Treeify supports MCP integration with enterprise workflows, systems, and AI pipelines. If you want to integrate Treeify into your testing platform, internal enterprise system, or AI workflow, you can refer to the following documentation:
http://8.130.27.55:3000/docs/cn/4-concept/4-2-mcp
Q24. Our company has strict information security requirements. What should we do?
If your organization has high information security requirements, you can choose Treeify’s private deployment option.
For more information, you can contact us on WeChat: TreeifyAI
6. Pricing, Feedback, and Contact
Q25. How are credits charged?
Treeify credits are fundamentally based on the token consumption of the underlying large language models.
To help protect information security, we use only official LLM APIs. These currently include:
- Azure OpenAI
- Anthropic Claude
As a result, credit consumption is tied to the actual token cost incurred during model usage.
Q26. How can I submit feedback to Treeify?
You can submit suggestions through the in-product feedback entry, or email us directly at:
contact@treeifyai.com
We highly value user experience and feedback, and we incorporate it into our product iteration process.
Q27. How can I contact Treeify on WeChat?
You can contact us on WeChat: TreeifyAI
If you would like to learn more about Treeify, you are also welcome to visit our website: