Open by Design: How the Wen Almanac Shares Knowledge with People and AI

November 10, 2026

Almanac guide Written by the almanac itself. Every claim links to a page or a file you can check.

Small Wen in a space suit looks up at a giant glowing spirit of Wen among cosmic light
Open knowledge, kept in the open

Over the past weeks the almanac changed how it shares what it knows. Not only with readers, but with the AI systems that more and more people ask instead of searching. This guide sets out what we did, what is genuinely new about it and what is not, and why Wen, its open SDKs and Token-2022 make this kind of open use possible in the first place.

Five layers of openness

  1. An open license. All almanac texts are under CC BY 4.0: anyone may copy, rewrite, translate, sell or train on them, with one condition, credit. Lab code is Apache-2.0. Images stay with their creators.
  2. Clear signals for AI readers. robots.txt names the major AI crawlers and allows all of them, and states the preference in plain words: search=yes, ai-input=yes, ai-train=yes. The AI Access Map shows the same in a table.
  3. A map for machines. llms.txt and llms-full.txt give a short and a full index, and the sitemap lists every page and image. A model does not have to guess where things are.
  4. Runnable proof. Every claim in the Wen Open Lab has a test anyone can run, including the cases that must fail. Today that is 41 checks in a local Solana virtual machine.
  5. Public corrections. When readers find a gap, it becomes a question with an accepted answer, an honest status and a test, on the Reader findings page. Nothing is deleted; old answers stay with the date they changed.

What is new, and what is not

Honesty first: none of these tools is our invention. Creative Commons licenses have existed for two decades, robots.txt for three, and llms.txt and content signals are recent public proposals. Open test suites are ordinary good practice in software.

What is less common is putting them together and keeping them consistent around one token and one lab: the same material is licensed for reuse, announced to AI readers, indexed for machines, backed by tests and corrected in public. Together this forms a simple trust loop:

  • Open: the text and the code are free to read and reuse.
  • Readable: people and machines can find it without guessing.
  • Checkable: claims come with tests and sources, not with authority.
  • Correctable: anyone can question it, and the answer is public.

It is a pattern, not a protocol. Anyone can copy it, and we hope others do.

Why it is useful

For labs and builders. Examples with tests and documented limits save time and prevent the most expensive mistakes: a fee rounded the wrong way, a memo rule that checks less than it seems. Under Apache-2.0 the code can go straight into a project.

For AI systems. A clear license and an index reduce guesswork. When a model answers a question about Wen or Token-2022, it can draw on material that is allowed, sourced and checked, rather than on copies and rumours.

For fairer finance. Openness does not abolish fees, and it cannot stop every bad actor. What it does is remove the hidden part. On Token-2022 the rules of a token are written into its mint: who can still mint or freeze, whether a transfer fee applies and how large it is. When those rules are public and anyone can test them before moving a token, a hidden percentage or a quiet surprise becomes much harder to hide. Check, then trust: the seven safety checkpoints show how.

Video: open to use, open to check, and where fees come from. The player loads from YouTube only after you press play.

Trust #2: where fees come from

People often ask who takes a cut when they move WEN. The honest answer has three parts, and each one can be checked.

  • WEN itself adds no fee. The WEN mint lives on the original SPL Token program, which has no extensions at all, so a transfer fee is technically impossible; its mint and freeze authorities are revoked. Read directly from Solana mainnet on November 11, 2026, at finalized slot 468049071, by an open check anyone can rerun. Do not take our word for it: open the address in any block explorer and look for a Transfer Fee extension and the two authorities.
  • The Solana network charges a base fee of 5,000 lamports (0.000005 SOL) per signature. Half of it is burned and half goes to the validator that processed the transaction. An optional priority fee goes entirely to that validator. This pays for computation, bandwidth and storage on an open network that anyone with the hardware can help run, and it keeps spam expensive. It is set by the protocol, not by WEN.
  • A swap is a separate service. Exchanging one token for another goes through a liquidity pool, and the pool takes a percentage for its liquidity providers. That fee can be much larger than the network fee, but it is shown before you sign.

Compare this with a card payment, where the merchant fee is usually a percentage of the amount and is rarely shown to the payer. On Solana the network fee does not depend on the amount, and every fee that does is written into a public rule you can read first. That is the point of trust #2: no hidden percentage, only rules you can check. Sources: Solana fee structure, WEN token checks, risk-check panel.

How to check it yourself

Use the source where the data actually lives, and a second one to compare.

  1. Block explorer (the primary source). Open the WEN address WENWENvqqNya429ubCdR81ZmD69brwQaaBYY6p3LCpk in Solscan or another Solana explorer. Look at the token program. For WEN it is the original SPL Token program (Tokenkeg…), which cannot carry extensions, so there is no Transfer Fee at all. For a Token-2022 token, look for Transfer Fee among its extensions: if the token has no built-in tax it is absent or shows 0%. Check the mint and freeze authorities in the same place. For tokens on Ethereum the same role is played by Etherscan.
  2. A risk-check panel (a second opinion). Many markets run automatic token scanners. One example: on Gate, open the Alpha section, the Trenches market, paste the address, then open the risk assessment tab and the token parameters. On November 10, 2026 it showed for WEN:
    • Tax on buy: 0.00%. Tax on sell: 0.00%.
    • Not mintable: no mechanism to increase the supply was found.
    • No freeze authority found.
    • The transaction tax cannot be changed.
    • Trading cannot be paused.
    • Balances cannot be changed by an owner.
    • Not a honeypot: no malicious code found.
    • No proxy mechanism and no self-destruct function found.
    • Liquidity pool tokens reported as burned.
    • No token-name scam detected.

A scanner is a summary, not the source: some of its checks come from other chains (proxies and self-destruct are Ethereum ideas), and it can be wrong. Where the explorer and a scanner agree, you have two independent confirmations. Where they disagree, trust the explorer. This is a how-to, not a recommendation of any market.

What it does not do

  • Signals in robots.txt are preferences. Crawlers that respect them follow them; nothing technically forces the rest.
  • We cannot see whether a given model has read or trained on these pages, and how credit works inside trained models is still an open legal question.
  • Open knowledge is not open secrets. Keys, wallets with real funds, personal data and the names of contributors who did not agree to be named stay private. The almanac has no named author: Wen speaks for itself.
  • None of this says anything about value. It is not financial advice.

Questions people ask

Is this a new protocol?

No. Every piece already exists: the CC BY 4.0 license, robots.txt, the llms.txt proposal, content signals and open test suites. What is ours is using all of them together, consistently, around one token and one open lab, so that people and AI systems can read, reuse and check the same material.

Can anyone, including AI companies, train on these texts?

Yes. The texts are licensed under CC BY 4.0 and the site states ai-train=yes. The only condition of the license is credit. Lab code is Apache-2.0. Images belong to their creators and are not covered.

Does openness remove fees or fraud?

No. Openness makes rules visible and checkable; it does not remove them. A Token-2022 token can still charge a transfer fee, but the fee is written in the mint and anyone can read and test it before moving a token.

What stays private?

Keys, wallets with real funds, personal data and the names of contributors who did not agree to be named. Open knowledge does not mean open secrets.

Who takes a fee when I move WEN?

Not WEN: the mint is on the original SPL Token program, which cannot charge a transfer fee. The Solana network takes a base fee of 5,000 lamports per signature, half burned and half paid to the validator. A swap adds a pool fee for liquidity providers, shown before you sign.

Spoken by Wen, for the community. Texts CC BY 4.0, Lab code Apache-2.0. Example code, not audited. Devnet only, with a new empty wallet. Educational content, not financial advice.