Study

Turn material into understanding the work can act on. The material may already be at hand, gathered earlier, or found during the task; the useful result is an explanation, a mechanism relation, a real disagreement, or a new judgment, not a pile of per-source summaries that still needs to be combined later.

Establish what the understanding must serve

Recover the desired outcome, the decisions this understanding will support, the material already present, and the authority granted. For an active project, frontload project inquiry: inspect its accepted purpose, current artifacts and use, accepted constraints, and live proposals before choosing what the inquiry should be. Note important targets and any counterexamples or alternatives the person wants covered, so the result answers the questions that matter instead of every question the material happens to allow.

When the person asks what a project should pursue next, the missing narrower question is part of the commission. Discover it from the project's accepted outcome, artifacts, and use rather than from a generic repository tour, and continue the same work without waiting for the question to be supplied. Ordinary repository investigation belongs with the active engineering method when its question is already clear.

Understanding is often a complete deliverable on its own. Do not defer the synthesis, hand the person the pieces to conclude from, or let the work settle into "read these and summarize".

An unresolved relationship in existing material can supply the next worthwhile question. Explain why answering it would change understanding or making, and pursue it within the commission. Preserve the user's open horizon; a promising question does not become the only permissible thesis. When the request is to continue, grow, or sound out an existing project, read project-grounded sounding. Naming promising questions is intermediate work, not the result.

If the discovered question turns on an external possibility, inspect it under the active task's authority. Read proactive inquiry and reference choice when the missing understanding depends on a substantial open choice and what implemented systems, experiences, mechanisms, or adoption conditions can show. Use enough material for the judgment rather than settling for descriptions of the surface.

Reconcile the materials

Read across the material for what each piece actually establishes and how the pieces relate: where they agree, where they conflict, and where they are really describing different objects. Align definitions, populations, versions, and conditions before treating two results as comparable; much apparent disagreement is scope, not substance. Read aligning-materials when sources use the same word for different things, when the amount of independent evidence matters, or when source fact, inference, and transfer must stay apart.

Distinguish a genuine disagreement, where the same question is answered differently, from a difference in what was asked. When materials truly disagree, explain what would settle it and whether the task needs that resolution now.

Form the explanation

Reconciling the materials is not the endpoint; the study should leave you with an explanation of your own. Propose the mechanism or relationship that accounts for what the material shows, or set two or three explanations against each other when the evidence allows more than one. Then test it: which cases it covers, which counterexample or negative result it leaves standing, and whether it still holds when a condition changes. A synthesis that only classifies agreement and disagreement has stopped short of the judgment the work can use. Read building-the-explanation when the material is rich enough to explain rather than only compare.

Fill only the gaps that matter

Material already present does not force a new search. Use what is there when it can support the judgment, and look outward when the commission itself asks for possibilities or when a decision-relevant gap remains and the current request permits it. Discovery can be useful without an already-known defect. If important material cannot be reached, name the specific gap and how it bounds the conclusion. Read coverage-and-limits when deciding how much more to gather, when coverage is partial, or when a conclusion risks reading stronger than the evidence.

Deliver the judgment

State what the material supports, what you infer from it, and what would be needed to carry a finding into this context. Surface the mechanism or relationship, not only the outcome, and keep a useful counterexample visible instead of smoothing it away. If a discovered project question produced an explanation, return it to the project's original use scene. Study does not need a unified score, a mandatory conclusion format, a fixed source count, or a new store of truth that the material does not justify.

Stay within the owner's boundary

Study explains and connects. It does not replace a specialized method's evidence standard and does not decide a product or value choice that belongs to the user. When material is present but its meaning must become a consequential decision, Shape owns that move; when the missing piece is an external fact, Search does. Use shared working context when another Skill or worker continues the same line of work.