Four ways to improve the work

Start with a claim you can improve

Choose a specific record, conclusion, definition or implementation behavior. Say what should change, why, and which evidence would let someone else assess it. A small correction with clear provenance can be more useful than a large unexplained rewrite.

The following paths describe the material a reviewer needs and the questions the review should answer. They are contribution guidance for this draft; a public submission process and contributor register have not yet opened.

Evidence

Bring a missing source, stronger corroboration, corrected technical documentation or a contradiction to an existing claim.

Include

• A source location, title, author or publisher, date and exact passage or artifact relevant to the claim.
• The affected claim or record ID, and whether the source supports, narrows or contradicts it.
• Retrieval context, version and any limitations on inspection or reuse. Explain whether accounts share an originating source.

Review asks

Can the source be inspected? Is the passage represented faithfully? Does it support this exact claim in this time and context? Stronger evidence may change an interpretation; source count alone does not establish independent confirmation.

Analysis

Offer a new threat case, a different maneuver decomposition or a competing causal explanation for the same evidence.

Include

• A bounded case and source set; separate reported behavior from your inference.
• Proposed conditions, mechanism and result, with stable references to affected records.
• The alternative's consequences: which conclusion or defensive placement changes, and what evidence could distinguish the alternatives.

Review asks

Does the model preserve the source's constraints? Are required conditions supported rather than assumed? Does it expose missing behavior? A clean graph is not sufficient evidence that the whole chain is executable.

Theory

Supply a counterexample, formal critique, semantic proposal, revised definition or testable hypothesis.

Include

• The exact specification or ontology version and the definition or contract being challenged.
• A minimal example showing the problem, plus the proposed interpretation or change.
• Consequences for existing records, compatibility, decision scope and the evidence needed to choose between alternatives.

Review asks

Is the issue a violation of an accepted contract or a proposal to change that contract? Does the change resolve the counterexample without introducing a contradiction elsewhere? A wording edit or review recommendation does not accept a semantic proposal.

The stronger coherent-witness support policy and the relationships between MC implementation variants and external Software concepts remain unresolved. Contributions should retain those open questions rather than silently select a replacement mapping.

Tooling

Improve an importer, validator, visualization, library or reproducibility workflow.

Include

• A small reproducible input, exact dependency and contract versions, and the expected behavior.
• A comparison with actual behavior, preserving failures and qualified outputs.
• Tests appropriate to the claim, known limitations, and a clear statement of whether the tool checks syntax, semantics, admission or evidence.

Review asks

Can another person reproduce the behavior? Does the transformation preserve meaning and provenance? Are failure conditions visible? A parser's success does not prove semantic conformance, and an implementation's speed does not establish practical security value.

Submission channel and decision responsibility

The intended public channel is a project email alias; its address has not been announced. The website repository was verified private on 19 October 2026 and its issue tracker is not a public submission path. Prepare the material above and retain it until an actionable contact channel is published.

Foundational decisions are currently author-led, with Jacob Schorr retaining responsibility for the core model. Proposed contributions require a scoped review; acceptance of material and acceptance of a new semantic policy are distinct decisions.

• Check contact availability
• Read the principles and governance direction
• Work from the available artifacts