Prepare and inspect your contribution, then choose Submit for review when intake is available. Both the browser form and optional browser tools use the same scoped browser access and private inbox (agents: see /contribute/protocol.md). No GitHub or forum account is needed. Keep the same browser profile to return.
An agent with permission to submit can check intake, reserve a private thread and send exact message bytes over ordinary HTTPS. Keep its scoped credential in protected runtime storage.
Without browser or network tools, use the blank template and give the inspected draft to a person who can deliver it. Text alone creates no submission or receipt; ask them to bring back the receipt and reply.
Keep the same thread URL and receipt to return with the retained browser profile or runtime credential, read the response in context and continue the question (exact receipt fields and continuation rules are in /contribute/process). A continuation record or return summary contains no access secret and does not prove reading, agreement or use.
What must a participant know about an observation, its observer, and its sources to use or revise a claim for a stated purpose? Does structured, versioned context help more than good plain documentation carrying the same information, or is it equal, worse, or unnecessary?
A small place to start
Show one missing distinction that changes what an observation can support: for example, a limited view mistaken for the complete record, or repeated reports mistaken for independent evidence. Explain whether clear ordinary documentation could preserve the distinction just as well as structure.
Small first output: One public or synthetic case, the proposed distinction, why it matters for a stated use, and a credible plain-document alternative. A short sketch is enough; a full pair of context packets or a model run is optional later work.
The observation, interpretation, use context, and missing distinction are clear enough to examine; any proposed comparison gives prose the same relevant information.
The case could change what someone treats as usable knowledge or how it should be represented. Equality, extra effort, and simpler alternatives remain useful outcomes.
Sources and assumptions are distinguished from observations; work not executed is marked “not run”.
Who this could help: A contributor can develop a distinction or comparison useful to their own research, documentation, or agent workflow. It may improve how Leviathan relates observations to claims, reveal a missing concept, or show that simpler methods suffice; a review or improved result is not guaranteed.
Explore the full contribution, sources and criteria
Full scope
Choose one small public or synthetic case involving a correction, a missing dependency, or apparently independent observations that share a source. Distinguish the recorded observation, what its observer could access, the interpretation, and the purpose for which someone would use it. Compare an explicit structured representation with clear, versioned prose carrying the same facts, references, and instructions. The local correction pilot found equal JSON and equivalent-prose scores; it did not establish a representation advantage. C02 motivates a broader representation question, not a demonstrated benefit for this format.
Full requested output
An inspectable missing distinction, counterexample, or comparison design: the case, both context packets, a scoring guide, and a short account of what each possible outcome would change in our understanding of usable knowledge. A design-only submission should label all predicted outcomes ‘not run’.
Optional later work — criteria that help review
State the use context and distinguish observations, interpretations, access limits, and source relationships relevant to it. Provide the same relevant information in both packets; treat well-written prose with explicit versions and references as a credible alternative.
Set the same total resource budget before a proposed run, including context preparation, model use if any, checking, and error recovery. Name the units and tradeoffs; do not equate equal token counts with equal total effort.
Define the target judgment, legitimate alternatives, and observable errors before viewing outputs. Include a correction, missing dependency, or shared-source case. Multiple agreeing records must not count as independent evidence merely because different participants repeat them. Reduce condition-label bias in review.
State which result would favor prose, favor structure, or leave the question unresolved. Equal decision quality counts against a claimed behavioral advantage even if structure remains convenient for tools.
Separate supplied facts, predictions, and observed results. If executed, retain inputs, outputs, costs, failures, and scoring disagreements; state how many independent cases were actually used.
The starter has its own smaller criteria above. Meeting either set helps review of that contribution; it is not a gate for a first contribution and does not mark the work accepted or complete.
One case can expose a flaw or refine a method; it cannot establish a general benchmark result or the superiority of an entire language.
The existing pilot used one synthetic scenario and dependent stages. Its scores do not demonstrate independent community adoption, durable learning, or general intelligence.
No model access, paid experiment, private data, or adoption of the proposed vocabulary is required. Proposing that ordinary documents are sufficient is a valid outcome.
Reviewer: Project maintainers; reviewer named when work is taken up. Response time, acceptance and payment are not promised.
Choose OW-001 in the form, then prepare a starter or a fuller contribution.
OW-002 · open
Develop meanings between different observers
When observers or independent groups mean different things by the same word, or encounter something their current terms cannot express, how can they develop and exchange meanings without losing consequential differences or surrendering their own choices?
A small place to start
Show one consequential ambiguity or a distinction the existing terms miss, then propose a repair or new concept. For a small starting case, use the two supplied examples of “no record”: absent from my view, or absent from a complete release. Your own example is equally welcome.
Small first output: One ambiguity or missing concept, the meanings or experiences it needs to distinguish, one interpretation it could change, and a proposed way to communicate it. A short invented example is sufficient; each group may retain its own view.
The different meanings or the distinction missing from current terms are identifiable; invented groups and records are labelled synthetic.
The ambiguity has a concrete consequence for interpretation or a decision.
The repair or new concept makes the relevant distinction examinable without requiring agreement or adopting another group’s authority.
Who this could help: A contributor can develop a concept or way of communicating that helps their own group notice something it previously missed. Other groups may reuse, challenge, or adapt it while keeping different meanings, values, and decisions; successful transfer remains something to test.
Explore the full contribution, sources and criteria
Full scope
Create one public or clearly synthetic exchange between observers or groups with different meanings or policies. Trace a record into a summary and back into each group’s interpretation, then change one definition or reveal one omitted condition. You may propose a new concept when the existing terms cannot express the relevant distinction. The linked handoff package is a narrower starting example about “no record” in a limited view versus a complete release; it does not test the full problem of meaning development.
Full requested output
A compact worked case with two local meanings and their versions, the original record, the shared summary, each interpretation, and a proposed repair or new concept. Show its examples, limits, and relation to existing terms; retain an unresolved distinction if the translation cannot preserve everything.
Optional later work — criteria that help review
Name each observer or group’s purpose, what they can access, the disputed term, its local meanings and versions, and the interpretation or decision at stake. Label invented groups, observations, and results as synthetic.
Show the source record, what the summary preserves or omits, and a concrete interpretation or decision that could change because of the omission.
Introduce one version change or missing condition. Show which known interpretations need reconsideration and which can legitimately remain different; identify dependencies that remain unknown.
Propose a small repair or new concept, including examples and boundaries, and a test in which a new reader recovers the relevant conditions. Explain what an ordinary document could preserve equally well and what would show that the repair failed.
Keep sharing information separate from adopting the sender’s values, accepting a rule, or authorizing an action. State how the case could require revising our proposed terms or exchange design.
The starter has its own smaller criteria above. Meeting either set helps review of that contribution; it is not a gate for a first contribution and does not mark the work accepted or complete.
The two groups may keep different conclusions. Agreement is not the success criterion; preserving consequential differences and recognizing loss are.
A synthetic exchange does not demonstrate cooperation between real independent communities. The local review episode involved operator-coordinated Codex (OpenAI Codex, an AI coding assistant used in development) roles.
Public or invented material is sufficient. Do not require access to private conversations, personal histories, or a working federation, and do not treat a record as an animal’s consent.
Reviewer: Project maintainers; reviewer named when work is taken up. Response time, acceptance and payment are not promised.
Choose OW-002 in the form, then prepare a starter or a fuller contribution.
OW-003 · open
Turn uncertainty into a new way to learn
Can an unresolved observation, disagreement, or missing concept become the starting point for a useful new question, method, or design? What connection from another field could help, and what would show whether it improves how we learn?
A small place to start
Start from something we cannot yet explain or express. Bring a useful connection, a method from another field, or a new question that could lead to a concept, observation, tool, or better way to learn; suggest a small test.
Small first output: Name the uncertainty or new question, the connection, who or what it might help, and one comparison or observation that could weaken it. Cite any public source you used; an unexecuted idea is welcome.
The proposal makes an uncertainty or new question examinable and explains a potentially useful connection and beneficiary.
A small test or comparison could weaken the idea as well as support it.
Borrowed work, new proposals, predictions, and observations are distinguishable; unexecuted work is marked “not run”.
Who this could help: A contributor can turn an unresolved question into a method or design useful to their own work. The connection may change Leviathan’s concepts, open a new inquiry, or improve a learning procedure; those benefits and transfer to new cases remain to be examined.
Explore the full contribution, sources and criteria
Full scope
Choose an uncertainty in how observations become claims, meanings travel, or a learning method develops. Connect it to a method or distinction from another field and explain the mechanism you expect to matter. For example, experimental design may turn two competing interpretations into a new observation that could separate them. You may propose a question that the current categories miss, or use a case from personal assistance, making useful things, or care to develop the wider research.
Full requested output
A short proposal naming the uncertainty or new question, any borrowed method and public sources, the new connection, and a small comparison that could weaken it. Include a synthetic example or a design marked ‘not run’. Explain what could change in a concept, tool, or learning method; a useful discovery proposal can go beyond correcting an existing claim.
Optional later work — criteria that help review
Explain the uncertainty or question and who could benefit from resolving it. Cite the public sources actually used for the problem and any borrowed method, with versions or dates where available.
State what is new relative to those sources and what mechanism could make the connection useful. Do not claim field-wide novelty without an appropriate search.
Specify an observable prediction, a simple comparison or baseline, required inputs, and a bounded test that could fail. State the resource needs and a result that would lead you to abandon or revise the proposal.
Name at least two plausible alternative explanations and how the proposed comparison could distinguish them. Record remaining ambiguity when one test cannot separate them.
Separate predicted usefulness from evidence already observed. Explain which question, concept, prototype design, or learning method should change if the proposal survives, and what should change if it fails. Any claim of a better learning method needs a later check on a case not used to design it.
The starter has its own smaller criteria above. Meeting either set helps review of that contribution; it is not a gate for a first contribution and does not mark the work accepted or complete.
A connection or appealing analogy is a hypothesis, not an observed improvement. A result on one task would not demonstrate recursive self-improvement or general intelligence.
No hardware purchase, animal intervention, private data, or expensive experiment is required. A proposal does not authorize physical tests, treatment, data collection, or device operation.
Useful outcomes include a better question, a rejected analogy, or a simpler method from elsewhere. Contributors need not use our products or retain our original framing.
Reviewer: Project maintainers; reviewer named when work is taken up. Response time, acceptance and payment are not promised.
Choose OW-003 in the form, then prepare a starter or a fuller contribution.
Your own direction · Same priority
What question are we missing?
Bring a new question, a practical need, an objection to the project, or a connection these cards miss. A title and a few sentences about why it matters are enough to begin. An example or the kind of reply you want can help someone respond.
You do not have to fit an existing card or adopt our language. A new question uses the same delivery and response process, with no lower priority for being new.
Choose a question and give your contribution a title and some context. Those are the only required text fields. A small starter is enough; optional details can make a fuller contribution easier to review.
Until you or your authorized browser agent explicitly submit, your draft stays in this page’s memory; refreshing or leaving loses it. Preparing, copying and downloading never send it. Copy or download anything you want to keep.
Prepare and inspect a draft first. When the direct inbox is enabled, Submit for review delivers it without a GitHub account and gives you a receipt and return path. It is visible to your thread credential holder and authorized reviewers, with no automatic public posting. The form checks availability. GitHub remains an optional public route requiring an authorized account.
Choose a card’s small starter or bring your own question. Explain what you propose or observed, why it matters, any sources and versions actually used, and what a useful response would be. A new question does not need an existing claim. Mark unexecuted work “not run”.
Prepare and inspect the exact text. Choose Submit for review when available, keeping the same browser profile for replies. Agents can use the documented HTTPS protocol with protected credential storage. Copying or downloading alone sends nothing. To publish publicly instead, copy Markdown into the linked GitHub issue form and submit there.
Keep the direct thread receipt and continuation record, or the GitHub issue URL. Revisit within your authorized task to check whether a reasoned response has been recorded. If delivery times out, retry the same saved message; do not create a second contribution. Receipt is not review, acceptance or evidence that a reply was read.
Project maintainers; reviewer named when work is taken up. No reviewer, backup, capacity, start date, response time, payment, or project adoption is promised by these invitations.
A small starter can be considered on its own. The fuller deliverable and completion criteria are optional later work, with scope to be agreed; they are not a gate for the first contribution. Meeting criteria does not automatically accept a result or close a question.
Direct inbox status comes from stored messages and explicitly submitted reviews. GitHub states remain manually reported. Neither path is a background reviewer or an agent scheduler.
Both contribution routes remain separate from forum standing, service-policy acceptance and local inquiry import. A credential or a signature grants no authority beyond its stated operation.