> For the complete documentation index, see [llms.txt](https://funarchy.gitbook.io/funarchy/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://funarchy.gitbook.io/funarchy/security-for-prediction-market/configuration/rulebook-risks.md).

# RuleBook Risks

### Ambiguous Resolution Criteria

#### Description

<figure><img src="https://4210179539-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2DiVEbUgCTsp2iPassR9%2Fuploads%2F1fmkiBZ96KsZU3qjPc4o%2Fimage.png?alt=media&amp;token=61c4561c-6397-441b-83ae-35ea4412ddd2" alt="" width="563"><figcaption><p>&#x3C;Example : Polymarket Rules></p></figcaption></figure>

The ambiguity of Resolution Criteria is not merely a planning oversight, but a structural vulnerability that enables profit-seeking participants to carry out ‘Semantic Attacks’.

When conditions are incomplete, creating gray areas in interpretation, participants that facing potential losses exploit these gaps inevitably. Regardless of objective facts, they persistently provoke disputes based on ‘technical definitions’ favorable to themselves.

In this case, the oracle becomes embroiled in a ‘logical battle’ rather than the ‘truth,’ creating a critical attack vector that compromises the integrity of the outcome and seizes the assets of the legitimate winner.

#### Real World Case Study

<figure><img src="https://4210179539-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2DiVEbUgCTsp2iPassR9%2Fuploads%2FK3dLv4mhW3swHOswqedp%2Fimage.png?alt=media&amp;token=52271165-40b7-4bf2-ac1c-cfcbb569e51e" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
[**Polymarket Zelensky Issue of Wearing Suit**](https://www.brewmarkets.com/stories/2025/07/08/scandal-engulfs-polymarket-over-zelensky-suit)&#x20;
{% endhint %}

{% stepper %}
{% step %}
A market was opened asking, “Will Zelenskyy wear a suit before July?”. \
However, the criteria were ambiguous not only regarding the physical definition of a ‘suit’ but also concerning ‘what state should be considered a proven fact’.
{% endstep %}

{% step %}
When Zelensky appeared wearing a jacket, most media outlets reported he was “dressed in a suit,” and the public interpreted it as a ‘Yes.’ \
However, the UMA Governance argued that “there is no clear consensus among credible reports.”
{% endstep %}

{% step %}
Ultimately, Oracle declared that “the consensus in the reports was insufficient and therefore unproven,” settling the resolved with a final ‘No’.\
Consequently, many participants who had bet ‘Yes’ based on their own image of formal attire of suit and media reports lost substantial assets due to a simple difference in terminology definitions and judgment criteria.
{% endstep %}
{% endstepper %}

#### **Mitigation**

* **Providing standard templates and guidelines**
  * Like  `[ Yes / No ]`, `[ Index / Price-based ]` , `[ Sports / Game-based ]`  Standardized resolution condition templates are provided by category, and ambiguous expressions (e.g., “quite,” “appropriately,” “good,” “is that so?”) are managed via a prohibited terms list to enforce phrasing consistency.
* **Mandatory specification of objective and verifiable indicators**
  * All resolution conditions must explicitly specify a clear reference data source (URL or API) and numerical criteria, ensuring they are documented in a format that allows third parties to verify and reproduce the results identically.
* **Disclosure of Resolution Logic and Priority Interpretation Principles**
  * Document and publicly disclose in advance which interpretation principles will be applied first in the event of a dispute, and link to them in the Terms of Service and Market Description.\
    This minimizes uncertainty and maintains consistent judgment criteria throughout the dispute process.

***

### Manual Oracle Delay

#### Description

This threat arises when the market's resolution criteria rely on ‘incomplete indicators’ such as a single website or manual updates by specific individuals.

While events in the real world have already concluded, a gap emerges between ‘the truth of reality’ and ‘the truth of the Oracle’ when the data sources that should reflect this (website administrators, announcements) delay updates due to administrative negligence or time zone differences.

This discrepancy isn't simply a delay in operations; it provides an attacker with a risk-free arbitrage window where they can "see the answer sheet and fill in their answers." This suggests that even with perfect market timing, a system can be compromised if the data source it references is corrupted or delayed.

Furthermore, structures that rely on a single point of contact are exposed to external attack vectors such as:

* **Insider Attack**: Deliberately delaying updates through bribery or collusion with staff.
* **Infrastructure Attack**: Physically blocking website updates through DNS hijacking or account takeover.

#### Real World Case

<figure><img src="https://4210179539-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2DiVEbUgCTsp2iPassR9%2Fuploads%2FshbpwuUfavaLUQVqdwC4%2Fimage.png?alt=media&amp;token=577d3c65-c927-4357-b860-18d1382f2b27" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
[**Polymarket OPM Shutdown issue**](https://thedefiant.io/newsletter/defi-daily/end-of-gov-t-shutdown-brings-more-polymarket-drama)
{% endhint %}

{% stepper %}
{% step %}
There was a market on the end of the US federal government shutdown, and the market was designed to determine the outcome based on the official update time from the Office of Personnel Management (OPM).
{% endstep %}

{% step %}
The actual shutdown ended on the 12th, but due to circumstances attributable to OPM officials, the website update was delayed and the update was released on the 13th. \
This delay created a one-day gap between the actual end time and the market's perceived end time.
{% endstep %}

{% step %}
Polymarket settled its markets based on the 13th, the date the website was updated, as required by regulations.\
As a result, even users who accurately predicted the outcome suffered losses due to simple administrative delays.
{% endstep %}
{% endstepper %}

#### **Mitigation**

* **Rulebook-Level Event Definition**
  * Instead of basing the settlement criteria on the "update time" of a specific Web2 page, the actual time of the incident (T) should be the standard of truth. \
    \
    Web2 pages are merely reference materials providing evidence, and should be designed to prevent operational factors such as update delays or human error from distorting the settlement results.
* **Evidence-Based Resolution**
  * The Rulebook should adopt an evidence-based settlement structure that does not rely on a single Web2 site as the standard of truth, but rather collects multiple independent pieces of evidence and verifies them through a community or DAO. \
    \
    To prevent manual updates to a single page from determining the settlement criteria, the final outcome should be determined based on multiple pieces of evidence.
* **Rulebook Transparency & Predictability**
  * The rulebook should clearly prioritize the factors that determine truth. In the event of a Web2 \
    page being delayed or not updated, the criteria for settlement and which criteria take precedence should be clearly defined in advance to ensure predictability and fairness.

***

### Use market picks that align with market results

#### Description

Depending on the market theme, results may be integrated, or the two may operate together.\
\
If the rulebook and market selection do not agree, the actual conclusion and selection range will be strictly applied, or conflicts may arise during the settlement phase, and the results cannot be excluded.

\
Therefore, the market design and selections that contradict the theme's outcome structure should be structured in a way that opposes them, and the rulebook should include settlement themes and backup cases.

#### Attack Scenario

<figure><img src="https://4210179539-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2DiVEbUgCTsp2iPassR9%2Fuploads%2F137SJQHvfCtt7Lev46gw%2Fimage.png?alt=media&amp;token=8b01e660-1ad7-4d5d-b4f8-908d30c20b8b" alt=""><figcaption></figcaption></figure>

{% stepper %}
{% step %}
A market has been launched to predict who will be selected as TIME's 'Person of the Year' and appear on the cover this year.
{% endstep %}

{% step %}
The market selection is structured so that only one candidate can be selected from several candidates.
{% endstep %}

{% step %}
TIME's cover sometimes featured multiple people rather than just one.
{% endstep %}

{% step %}
The market choices were designed to allow only one choice, so there were no choices that would allow for multiple correct answers.
{% endstep %}

{% step %}
The actual result should have been treated as multiple correct answers, but the market settled on the "Others" option. As a result, even those who had partially guessed the person on the cover were marked as incorrect and suffered losses.
{% endstep %}
{% endstepper %}

#### Mitigation

* It should be designed in advance to allow for flexibility in responding by including the ability to modify options during market progress depending on the results.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://funarchy.gitbook.io/funarchy/security-for-prediction-market/configuration/rulebook-risks.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.