frame

Read the discovery protocol before acting. Bundled templates are optional aids, not a mandatory artifact checklist.

Frame the real need

Read the supplied idea and available context. Identify the actors, progress sought, current workaround, desired observable outcome, status-quo consequence, supported constraints, decision criteria, and material unknowns. Preserve the distinction between a proposed feature and the underlying need, including technical needs.

Reuse what the user already told you. Ask a small group of consequential questions, not a form-filling interview. Label unverified observations as attributed reports or assumptions. Do not invent customer interviews or constraints.

For quick discovery use the Need section of discovery.md; otherwise save frame.md only when useful. Surface which question would change the next step. Normally hand off to /brainstorm <frame>; focused research or an already clear decision may be more useful. New evidence can return here without erasing earlier context. Do not choose a solution, approve a promise, or start implementation.

Compound durable learning

If this work surfaces a durable, non-obvious lesson likely to help future ITP work, invoke the compound-learning skill before finishing. Do not invoke it for routine facts, one-off details, already-recorded evidence, or merely because the work was difficult. Compounding is advisory: it must not create or replace a human decision, change an approved promise, add scope, or silently rewrite shared ITP behavior.