R6
Medium confidence
Proposed
Get quoted for a query you already rank for
One receipt from evidence through decision, application, verification, and outcome.
Priority
6.29
impact 52.4 × confidence 60% ÷ effort 5
Decision receipt
Why this fires
The attached observations crossed this rule's declared threshold. Fixing it changes the named target; measurement starts only after verification.
Observed search result
Answer-engine receipt
- Target page
- https://larkspur-roasters.example/guides/pour-over
- Target query
- best beans for pour over
- Impact estimate
- 52.4
- Measurement window
- 42 days after verification
1Observed search result
{"keyword": "best beans for pour over", "rank": 4}2Answer-engine receipt
{"checks": 4, "cited": false, "prompt": "best beans for pour over"}Options considered
Choose the road, not just the recommendation.
Implementation handoff
Artifact
Metered generation
Priced before it runs
A content brief is drafted by a language model against this action's evidence, so it costs credits. A workspace shows the credit estimate and asks before anything is dispatched. The sample never runs one, so there is no draft to show.
Selected approach: Answer the question in the page's own words, near the top, in one passage an engine can lift.
- Vendor dispatch
- Only after confirmation
- In this sample
- Never
Before and after
Outcome
An outcome appears after the verified measurement window closes.