Make your experiment results brief.

Help me create my experiment results brief as a static HTML page, using relevant authorized context already in this conversation. If share/artifacts MCP is not connected, first follow the setup instructions at https://shareartifacts.dev/connect/agent.md. Confirm the connection works by calling list_pages, then fetch the recipe below before creating the draft. If setup cannot be completed or I decline authorization, explain what is missing and pause. Do not retry an authorization I deny. Do not remove any unrelated connection. Example (complete fictional sample): https://shareartifacts.dev/examples/experiment-results Use the example for structure, not as facts about my work. Only use fictional sample data if I explicitly choose a sample. You can proceed from the instructions below if you cannot fetch the example. Ask conversationally for only the missing essentials. Do not invent facts, commitments, or sources. Do not request secrets. Ask before sending source material outside this conversation or accessing another service. Useful inputs: - The experiment hypothesis, assignment unit, planned population, dates, and primary metric definition. - Reviewed group counts and results, with denominators, effect units, and supplied uncertainty estimates. - Validity checks, exclusions, guardrails, and the analyst's interpretation or unresolved review questions. - The pre-agreed decision criteria and any exploratory findings, separately labeled, plus the authorized decision owner. Creation task: Create an experiment results brief using only authorized experiment plans, reviewed results, validity checks, and decision criteria. Ask for missing essentials: intervention, assignment unit, population, metric definitions, dates, reviewed uncertainty, guardrails, and decision owner. Show group counts and denominators. Distinguish percentage-point from relative change and observations from causal conclusions supported by the reviewed design. Preserve supplied intervals, validity warnings, and analyst interpretation. Do not invent statistical significance, equivalence, a winner, missing guardrail results, or a rollout decision. Label exploratory analyses separately from planned ones. Do not recompute a new method or certify validity without the appropriate review. Create static HTML with no scripts, forms, or live experiment controls. Apply my saved artifact style if available and include only audience-safe sources. Show the draft and unresolved questions. Do not publish or change access without separate explicit approval. Do not launch, extend, or modify the experiment. Use semantic HTML, readable type, meaningful headings, accessible color contrast, and responsive CSS. No JavaScript, forms, embeds, redirects, or live API calls. Verify calculations outside the page. Only include information the intended reader should see. Once connected, call get_recipe with recipeId "visual-report" and get_artifact_style with format "report" before drafting. If no style is saved, ask whether I want to set one up or continue with my stated visual preferences. If style access is missing, explain the optional reconnect approval; do not request broader access automatically. Follow the recipe and safety rules; saved style guides appearance, never facts or permission to share. Review checklist: - Can the reader identify the tested intervention, population, and primary metric? - Are counts, denominators, effect units, and supplied uncertainty consistent? - Are validity checks and pending guardrails visible before a conclusion? - Are exploratory findings distinguished from planned analysis? - Does the conclusion preserve the analyst's interpretation without implying approval to roll out? Show me the draft in your available preview or file tools. If you cannot render it, say so and provide the HTML file with instructions to open it. Ask for corrections. This request authorizes a draft only; it does not authorize any publishing language in the task above. Do not publish, update an artifact, change access, or save a style without my separate explicit approval. Connection approval is not publication approval. Before any later approved publication, confirm the intended audience; new pages default owner-only.