grill-with-docs

<what-to-do>

Interview me about the unresolved decisions in this plan until we reach a shared understanding. Map which decisions depend on others. For each question, provide your recommended answer and a short reason grounded in the project's code, terminology, or documented trade-offs.

Work in short rounds:

1. Resolve facts from the codebase and existing documents before asking the user. An unfinished lookup leaves its dependent decisions unready; other independent questions can still proceed.
2. Ask ready, independent questions together in a numbered round. A question is ready when its prerequisites are settled and its wording, options, and recommendation do not depend on another open answer. Usually two or three concise questions fit; ask one when it is the only ready decision or needs focused discussion. Honor a request for one-at-a-time questions.
3. Wait for feedback before asking dependent follow-ups. Apply each explicit answer, capture resolved terms as described below, and recompute which decisions are ready. With a partial reply, keep unanswered questions open; silence does not accept a recommendation. If the reply is ambiguous about which question it answers (for example, a bare “yes” after two different questions), clarify only that mapping instead of applying it to every recommendation. Clear answers need no numbered format; “yes to both” can settle both. Do not repeat settled questions unless new evidence changes their basis.
4. Finish when the scoped decisions are resolved, or name the remaining dependency if progress requires missing evidence or a user decision. Batching questions does not authorize implementation or turn a proposed answer into an accepted ADR.

</what-to-do>

<supporting-info>

Domain awareness

During codebase exploration, also look for existing documentation:

File structure

Most repos have a single context:

/
├── CONTEXT.md
├── docs/
│   └── adr/
│       ├── 0001-event-sourced-orders.md
│       └── 0002-postgres-for-write-model.md
└── src/

If a CONTEXT-MAP.md exists at the root, the repo has multiple contexts. The map points to where each one lives:

/
├── CONTEXT-MAP.md
├── docs/
│   └── adr/                          ← system-wide decisions
├── src/
│   ├── ordering/
│   │   ├── CONTEXT.md
│   │   └── docs/adr/                 ← context-specific decisions
│   └── billing/
│       ├── CONTEXT.md
│       └── docs/adr/

Create files lazily — only when you have something to write. If no CONTEXT.md exists, create one when the first term is resolved. If no docs/adr/ exists, create it when the first ADR is needed.

During the session

Keep answers attached to the decision

When several choices remain open, use the review and decision capture reference. It owns stable IDs, source versions, partial replies, and revalidation when the basis changes. Use the host's question UI or the existing project review page; an answer sheet does not replace the project's approval mechanism. Keep resolved answers in the existing plan or decision owner, rather than a second ledger.

Challenge against the glossary

When the user uses a term that conflicts with the existing language in CONTEXT.md, call it out immediately. "Your glossary defines 'cancellation' as X, but you seem to mean Y — which is it?"

Sharpen fuzzy language

When the user uses vague or overloaded terms, propose a precise canonical term. "You're saying 'account' — do you mean the Customer or the User? Those are different things."

Discuss concrete scenarios

When domain relationships are being discussed, stress-test them with specific scenarios. Invent scenarios that probe edge cases and force the user to be precise about the boundaries between concepts.

Cross-reference with code

When the user states how something works, check whether the code agrees. If you find a contradiction, surface it: "Your code cancels entire Orders, but you just said partial cancellation is possible — which is right?"

Update CONTEXT.md inline

When a term is resolved, update CONTEXT.md right there. Don't batch these up — capture them as they happen. Use the format in CONTEXT-FORMAT.md.

Don't couple CONTEXT.md to implementation details. Only include terms that are meaningful to domain experts.

Offer ADRs sparingly

Only offer to create an ADR when all three are true:

1. Hard to reverse — the cost of changing your mind later is meaningful
2. Surprising without context — a future reader will wonder "why did they do it this way?"
3. The result of a real trade-off — there were genuine alternatives and you picked one for specific reasons

If any of the three is missing, skip the ADR. Use the format in ADR-FORMAT.md.

</supporting-info>

Question-round sequencing adapted from Matt Pocock's grilling skill. Domain checks and inline documentation remain part of this skill.