> For the complete documentation index, see [llms.txt](https://docs.canhav.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.canhav.com/token-track-ideation/enforced-vs-stated.md).

# Enforced versus stated

**Available now** on public token design pages and in the studio.

Readers must be able to tell which claims are written into a contract at deployment and which are published commitments only.

## Labels

| Label             | Meaning                                                                                        |
| ----------------- | ---------------------------------------------------------------------------------------------- |
| Enforced on-chain | Written into the contract at deployment. Cannot be changed by anyone afterward for that token. |
| Stated by team    | Published in the design record (or journey). Not enforced by the token contract.               |

## What is enforced on-chain (CanHav launch token)

When you deploy through the CanHav factory:

* Fixed total supply minted at initialize (no later mint)
* No pause, freeze, or blacklist on the token
* No upgrade path and no owner on the token
* Source pre-verified because the factory clones a verified implementation
* Optional vesting wallet parameters you set in the same launch transaction (cliff, duration, percent)
* Launch fee paid at create time

Verify: [Contract guarantees](/token-launch/contract-guarantees.md) and the [TokenFactory](https://explorer.testnet.chain.robinhood.com/address/0x30Db3A828F65B92434c6aDB27AEeD01850277b08).

On a design deploy, the launch transaction also records the design snapshot hash as `journeyHash`, so the published design is tamper-evident relative to that launch. See [Deploy paths](/token-launch/deploy-paths.md).

## What is stated by the team only

These live in the design document / snapshot. They are not rewritten into immutable token bytecode as allocation tables:

* Allocation narrative (who gets what percent among team, investors, treasury, public, liquidity, advisors, other)
* Distribution plans (sale terms, undersubscription plan, airdrop intent)
* Token rationale and “beyond a database row”
* Market plans that are not executed in the launch transaction
* Governance and legal status declarations
* Post-launch runway and reporting intent

A sophisticated reader treats stated commitments as claims to check against later behaviour, not as contract guarantees.

## Status declarations

Legal, governance, and security status fields are always **stated by the team**. See [The three answer types](/getting-started/three-answer-types.md).

## Related

* [Contract guarantees](/token-launch/contract-guarantees.md)
* [Public pages](/public-pages/public-pages.md)
* [Token track](/token-track-ideation/token-track.md)
* [Deploy paths](/token-launch/deploy-paths.md)


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.canhav.com/token-track-ideation/enforced-vs-stated.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
