← Back
Frequently Asked Questions

Understand the account before you trust it.

CheckDragonfly is an AI-assisted account intelligence application designed to help you evaluate publicly available information about accounts on X. By combining account data, behavioral evidence, and in-depth AI analysis, CheckDragonfly helps turn complex information into a clearer, more structured assessment.

1. What is the CheckDragonfly application?

CheckDragonfly is an account intelligence and evaluation platform that helps you better understand an X account before deciding how much confidence to place in it.

You can check any public X account: your own handle, or anyone else’s you are about to trust, follow, or do business with. CheckDragonfly reads only what X makes public about that account.

An account’s profile, posting history, and public interactions can provide useful context, but reviewing those signals individually can be time-consuming and difficult. CheckDragonfly brings relevant information together and evaluates it through a structured analytical process.

The application combines available account data, evidence-based scoring, and AI-assisted analysis to help identify patterns, inconsistencies, credibility signals, and areas of uncertainty.

The goal is to help you make more informed decisions—not to ask you to accept an unexplained score.

This report is for informational and educational purposes only. It describes signals in public data and does not label, accuse or make claims about any person.

2. How does CheckDragonfly work?

CheckDragonfly uses a multi-step evaluation process:

  1. Account data collection: The application retrieves eligible account and post information through authorized API access, subject to the data available for the account and the applicable API permissions.
  2. Evidence analysis: Relevant account characteristics, posting activity, and available behavioral signals are organized for evaluation.
  3. AI-assisted account review: Grok performs a deeper analysis of the available account information, looking for patterns, consistency, context, and potential warning signs.
  4. Scoring and interpretation: Analytical results are combined into an overall assessment intended to make the available evidence easier to understand.
  5. Results presentation: The application presents its assessment and supporting information so you can consider both the findings and their limitations.

CheckDragonfly distinguishes between evidence that supports a conclusion, information that warrants further review, and information that remains unknown.

The depth of an evaluation depends on the information that can actually be accessed. Protected information, unavailable history, deleted content, and other data limitations can restrict what the application can conclude.

3. Why would CheckDragonfly be useful to me?

Online interactions increasingly influence purchasing decisions, professional relationships, business opportunities, investment discussions, and the information people choose to believe or share.

CheckDragonfly can help you examine an account more systematically when you encounter an unfamiliar person, creator, business representative, commentator, or other account on X.

Potential uses include:

CheckDragonfly is intended to supplement your judgment. It is not a substitute for independent verification when money, safety, reputation, or other important interests are at stake.

4. Why does an CheckDragonfly evaluation cost money?

A meaningful account evaluation involves more than retrieving a username or displaying a profile.

CheckDragonfly is designed to combine data retrieval, evidence processing, analytical scoring, and a deeper Grok-powered review. Each stage contributes to the work required to produce a useful assessment.

There are two principal cost drivers:

The cost of accessing account data. X’s Developer Platform charges for certain API resources, including user and post data. Its published pay-per-use pricing currently lists $0.010 per user read and $0.005 per post read. Actual costs depend on the resources retrieved, current pricing, and the scope of an evaluation. See the X Developer Platform.

The cost of deeper AI analysis. Grok is used for more than a simple text response. The evaluation process can involve examining multiple pieces of account information, interpreting relationships between signals, and producing a more comprehensive account-level assessment. AI processing costs depend on the model, the amount of input and output, and the analysis performed. See the xAI API documentation.

CheckDragonfly also requires software infrastructure, development, testing, data processing, and ongoing maintenance.

The price therefore reflects the overall evaluation process—not simply access to a public profile.

Choosing what to check. Each price lists up to three history windows: 30 days, 100 days and full history. A new account only gets the windows its age allows. Full history is limited to the 3,200 most recent posts X makes available. Post and picture counts for 100 days and full history are estimates from the account’s 30-day counts and profile totals. Each window has one price, and picture analysis is included in every window. Your report says which window it covered.

You never pay for pictures that could not be read. AI may not be able to read some pictures. Each picture in a check gets exactly one AI reading attempt, and the full cost of every picture that cannot be read, for any reason, sales tax included, is refunded automatically and immediately when the check finishes, to the card you paid with. You are emailed the amount. Banks typically take 5-10 business days to show the refund.

Are price checks free? Your first 3 price checks each day (Central Time) are free, and checking the same account again within 7 days is always free. After that, each extra price check adds what it cost us to read X’s data, usually about $0.03, at cost, to your next price, never more than $1.50, and the price window shows it. You pay it only if you buy that check. Daily limits apply: 30 price checks if you have bought a check before, 10 if not, 5 on your first day.

The value proposition is depth and structured analysis: bringing relevant information together so you can evaluate an account more efficiently and make a better-informed decision.

5. Why do I need CheckDragonfly?

You may not need an account evaluation for every interaction on X. CheckDragonfly is most useful when the credibility of an account matters to a decision you are about to make.

A profile can look convincing without establishing the accuracy of its claims. A large audience does not, by itself, prove credibility. A small audience does not, by itself, say anything about whether an account is authentic.

CheckDragonfly is designed to help you look beyond superficial signals by considering multiple pieces of available evidence together.

Before you trust a claim, engage in a business discussion, follow advice, or invest time in a relationship, an independent account evaluation can provide additional context.

CheckDragonfly cannot eliminate risk or guarantee that an account is genuine. It can, however, help organize the available evidence and highlight questions worth investigating before you act.

You make the decision. CheckDragonfly helps you examine the evidence.

6. Is CheckDragonfly legal?

CheckDragonfly is designed to evaluate eligible, publicly available X account information obtained through authorized API access. X provides a Developer Platform with documented API endpoints that developers can use to retrieve permitted account and post data for their applications. Grok can then be used as an AI analysis component through the separate xAI API.

Relevant official resources include:

Using an official API is an important part of building an application responsibly, but API availability alone does not automatically establish the legality of every possible use. Compliance also depends on the applicable developer agreement, platform policies, access permissions, privacy obligations, and the particular use of the data.

CheckDragonfly is intended to analyze permitted account information for informational and analytical purposes. It is not intended to access private account information without authorization, bypass platform restrictions, or represent AI-generated assessments as independently verified facts.

No automated evaluation can guarantee a person’s identity, intentions, or trustworthiness. Users should consider the results alongside other relevant information and should not treat an CheckDragonfly assessment as definitive proof of fraud, authenticity, or misconduct.

This explanation describes the application’s intended approach; it is not a legal opinion or a guarantee of compliance in every jurisdiction or circumstance.

7. Can CheckDragonfly guarantee that an account is real or trustworthy?

No. CheckDragonfly evaluates available evidence and identifies patterns that may support further investigation. It cannot conclusively establish a person’s real-world identity, intentions, honesty, or future conduct solely from an X account.

An account may have legitimate reasons for unusual activity, and an apparently consistent account may still make misleading claims.

Results should therefore be interpreted as an evidence-based assessment with limitations—not an absolute verdict.

8. What if there is not enough information about an account?

Not every account provides the same amount of accessible information. Some accounts have limited activity, and some data may be unavailable through the API or outside the scope of an evaluation.

When evidence is limited, the assessment should reflect that uncertainty rather than invent missing information or treat the absence of evidence as proof of suspicious behavior.

A useful evaluation is not one that pretends to know everything. It is one that makes clear what the available evidence can—and cannot—support.

9. Does CheckDragonfly replace my own research and judgment?

No. CheckDragonfly is a decision-support tool. It is designed to help you examine available information more systematically, but it does not replace independent research, direct verification, professional advice, or your own judgment.

For significant financial, professional, or personal decisions, independently verify important claims and consult appropriate sources.

10. Do AI-made pictures affect the score?

Yes. Each picture gets one AI reading for signs of being AI-made or AI-edited, and the profile picture is read too. A likely AI-made profile picture lowers the identity part; flagged post pictures lower the content part and the AI reading; Grok weighs them in its review. Pictures that couldn’t be read never count against an account (and are refunded). Pictures are also judged in context: disclosed or obvious AI art, memes and parody aren’t penalized, while an AI face presented as a real person lowers the score more, and the strongest finding needs several independent signs beyond the pictures. A flag is not proof of wrongdoing.

11. Do you use facial recognition?

No. CheckDragonfly never uses facial recognition and never identifies anyone from their face. To check for impersonation, it compares the account’s claimed name and handle with verified accounts, then checks whether its profile picture or banner is an exact or near-exact copy of that verified account’s image, and whether its bio is copied. Pictures are compared as whole images, using a small fingerprint of the image file, not by measuring faces, so a different photo of the same person is not matched. No biometric identifiers or templates are created.

12. Who is CheckDragonfly for?

CheckDragonfly is intended for people who want additional context about X accounts before relying on their claims or deciding how to engage with them.

Depending on the situation, that may include individuals, professionals, business owners, researchers, and people who regularly encounter unfamiliar accounts online.

The common need is straightforward: a clearer way to examine the evidence behind an account before placing confidence in it.

13. Can I get a refund?

Pictures that could not be read are refunded automatically when a check finishes. Other refunds are reviewed by the owner. The help chat can’t issue or file refunds: message the owner (the ? button, then message the owner) and they’ll check your paid checks against our records. A refund can be given when a paid check failed or never finished, you were charged twice or more than the quote, we checked a different account than the one you paid for, or the report is still missing sections after a re-run. A partial refund can be given when the account went protected, suspended or deleted before the check ran, or X returned much less history than the window you paid for. Disagreeing with a score or finding, a wrong account you typed and confirmed, or a change of mind after the report arrives are not refundable. See the Terms.

14. How does the help chat work?

Press the green ? button on any page. The help chat answers support questions (billing, refunds, accounts, reports, prices) in text; it has no voice and never uses your microphone. It never ticks boxes, submits forms or pays for you, and for anything it can’t answer you can message the owner.

Make a more informed decision.

You cannot eliminate every risk associated with online interactions. You can, however, take a more structured approach to evaluating the information available to you.

CheckDragonfly — account intelligence designed to help you look beyond the profile.

Visit checkdragonfly.com to learn more.

© 2026 CheckDragonfly. All rights reserved.