// 01 · what it does

Capture decisions that stay found.

decision-log has two modes. Capture extracts a decision from past context — a meeting summary, a Slack thread, a brief — and produces a structured decision record: decision statement, who decided, context, options considered, rationale, implications, and reversibility. Structure frames a pending decision: it defines the question precisely, maps the options and trade-offs, assesses reversibility, makes a recommendation if evidence permits, and names the decider and timeline.

Decisions are re-litigated because no one wrote them down, and when someone did write them down, the doc is unfindable six months later. decision-log solves both: it structures the record so it's useful when retrieved — not just what was decided but why, what was considered, and what would change the call. Every captured decision appends to a master DECISION-LOG.md in knowledge/decisions/, which means the full decision history is in one place.

Day 12 follows roadmap prioritization because the PM just made sequencing calls that should be captured while the reasoning is fresh. The skills in this week build on each other — the prioritization rationale (Day 11) informs the decision log (Day 12), which feeds the alignment memo (Day 13).

// 02 · sample prompts

Two ways in.

prompt.basic.txt
/decision-log

Capture this decision: our team decided last week to not build a native Android app this year and focus on making our mobile web experience better instead. The main reason was cost — native Android would take 6 months of eng time. Our lead engineer also argued that most of our users are on iOS anyway.
prompt.advanced.txt
/decision-log

Structure mode. This decision is pending and needs to be made before we brief the rollout expansion team next week.

Decision question: Should Terrain mandate Instant Book for all guides who meet the quality threshold (rating >4.7, 10+ completed bookings, no safety incidents in the past 12 months), or introduce a flexible Instant Book mode that lets eligible guides set a configurable minimum advance-notice window (e.g., 2 hours, 4 hours, 24 hours)?

Context: Instant Book is currently in beta with 15% of guides. Only 22% of beta-eligible guides have opted in — far below the 45% target. Detail-to-booking conversion on Instant Book listings is about 32% higher than request-to-book. The Day 8 data analysis confirmed this lift is directionally consistent across all experience categories. The primary guide objection (from Day 7 feedback synthesis and recent guide feedback) is calendar risk: guides worry about double-booking private clients and getting no-shows when they can't vet adventurers before confirming.

Options considered so far:
Option A — Mandate Instant Book for eligible guides: Would maximize conversion lift. Jordan Lee (Guide Experience PM) believes this is the right call. Risk: guides may churn or deactivate listings rather than comply.
Option B — Flexible Instant Book with advance-notice window: Addresses guide calendar concern directly. Reduces churn risk. Slower conversion lift (some guides may set long windows). Requires additional backend work to implement the window logic.
Option C — Continue voluntary opt-in with better education: Lowest build cost. Current trajectory suggests 22% opt-in is a ceiling, not a floor. Conversion impact deferred indefinitely.

Stakeholder positions: Jordan Lee (Option A), Marcus Webb / Legal (concerned about Option A liability exposure if mandatory adoption creates disputes), Priya Anand (wants faster conversion lift, prefers Option A or B). Riley Chen / Trust & Safety (flagged that forced opt-in could increase cancellation disputes if guides feel coerced).

Decider: Dana Park (VP Product).
Decision needed by: end of this week.
Reversibility: Option B is more reversible than Option A (can adjust the window logic; harder to un-mandate).
// 03 · reflection

Three questions.

  1. 01What did structuring the decision reveal about which option actually has stronger evidence behind it?
  2. 02Who does the skill identify as the actual decider — and are the right people involved before that decision gets made?
  3. 03What would you add to this decision record to make it genuinely useful to someone reading it 6 months from now?