← ClaudeAtlas

resolve-open-questionslisted

Use when a delivery run's planning has blocking or open questions that must be answered by the human before requirements can be treated as resolved: asking each focused required question, recording the answer in writing, and moving register rows from questioned to resolved. This is the resolve_open_questions action of the planning stretch — one Agent-tool dispatch with subagent_type business-analyst — distinct from gather-requirements (which raises the questions) and from the delivery-lead's own checkpoint escalations.
BhangeeF16/kaizen · ★ 0 · AI & Automation · score 65
Install: claude install-skill BhangeeF16/kaizen
# Resolve Open Questions Answer planning's blocking questions with the human, in writing — never on the requester's behalf. This action exists precisely so no later action inherits an unanswered material question. ## Dispatch One synchronous `Agent`-tool call, `subagent_type: business-analyst`. Because this action's whole job is interaction with the human, the dispatch prompt must state which questions are open and require the subagent to stop rather than infer an answer; when the run's own flow has the delivery-lead relaying answers, the answers relayed must be the human's actual words, not a paraphrase that resolves ambiguity by accident. ## Required inputs - The blocking question(s) — from the register, the target's status record, or an earlier action's report. ## Required outputs (all mandatory) - Every answered question recorded in `planning.md` (interactively with the user and in writing) and blocking questions cleared in the target's status record. - Every requirement-register row moved from questioned to resolved. ## Rules - One focused required question at a time beats a questionnaire — and no question may be marked answered without the answer recorded in writing. - A requirement whose conflict with an existing business rule or decision remains unresolved does not move to resolved — the conflict itself is the open question. ## Result Report each required output produced (with its location), the result (pass / pass_with_risks / blocked / failure), and block