Generate Flintmere experiment readouts from design and results exports.
npx skills add https://github.com/Flintmere/flintmere --skill experiment-readout
Please help me install this Agent Skill. Skill: experiment-readout Source: https://github.com/Flintmere/flintmere/tree/main/.claude/skills/experiment-readout Command: npx skills add https://github.com/Flintmere/flintmere --skill experiment-readout
This Skill enables teams to generate honest, rule-based readouts for completed Flintmere experiments, ensuring adherence to pre-declared decision rules, updating experiment logs, and surfacing actionable learnings.
Run the readout workflow after the experiment window closes to generate the final readout.
High-intent search queries and answers about installing and using this skill.
To generate an experiment readout, run the workflow after the experiment window closes. It identifies and validates outcomes using pre-declared designs and results exports, summarizing primary metrics, confidence intervals, and learnings.
Applying pre-declared decision rules ensures compliance when closing experiments. The readout workflow validates the final results against the original experimental design, articulating the outcome and updating the experiment-log entry.
Documenting experiment learnings involves computing primary metrics, confidence intervals, and effect sizes from results exports. The workflow then emits a formal readout document and updates project logs with actionable follow-ups.
Yes, you can update the experiment-log automatically. The readout workflow ensures compliance with the pre-declared rule, updates the experiment-log entry, and emits the final readout document in one step.
Yes, results exports are required to calculate effect sizes and primary metrics. The workflow validates outcomes by articulating results across the design references and the experiment-log entry to ensure an honest readout.
A pre-declared experimental design is required to ensure an honest, rule-based readout. The workflow validates the final results against this original design to determine the outcome and maintain compliance throughout the closure process.
MilUX-LtdDesign minimal falsifiable experiments with pre-registered outcomes and decision rules.
louisfghbvcOrganize project learnings across sessions in a local JSONL database.
nimrodfisherCompute statistical significance, confidence intervals, and effect sizes for A/B test variants.
amplitudeSummarize A/B test outcomes and recommend next steps for stakeholders.
