Evidence-first AI visibility workflow.
Use this playbook to connect prompt discovery, recurring monitors, answer evidence, citation repair, alerts, exports, and agent orchestration into one operating loop.
Brand mentioned in the answer body, not only repeated from the prompt.
Expand category and comparison prompts where the brand is absent.
Owned and trusted third-party sources support the answer.
Prioritize source fixes before treating score movement as durable.
The brand appears before competitors in shortlist and buying-intent answers.
Write comparison proof and objection handling for prompts competitors win.
Cited pages include current product, pricing, docs, and trust context.
Refresh answer-ready sections and validate crawler access after publishing.
Run the same prompt on competitors
One score is directional; comparison shows whether the category answer favors another brand and which prompt should become a recurring monitor.
Save recurring prompts as monitors
Turn high-value prompts into a daily monitor so answer changes become trend evidence. AI Overview prevalence varies materially by industry and query set — track whether you appear.
Export shareable proof
Use markdown, JSON, CSV, or PDF artifacts when the answer must reach stakeholders. Reports include citations, sentiment, and competitive context.
Run CLI agent orchestration
Use `npx prompts-gpt orchestrate --mode eval` to race agents, score results, and automate visibility workflows from your terminal.
Competitor overtake
Competitor appears ahead of the brand in 2+ saved prompts. Send to weekly report and mark as warning immediately.
Citation loss
Owned citation share drops below 25% or loses a key URL. Notify content owner with source and prompt context.
Sentiment drop
Negative or mixed answer framing appears on buying-intent prompts. Open a remediation brief before the next stakeholder report.
Evidence confidence
Fewer than 4 saved answers or fewer than 2 engines for a priority prompt. Keep the alert in watch state until the next scan validates the movement.
Executive
Score movement, competitor risk, business impact, and top three decisions.
SEO and Content
Prompt gaps, source quality, citation repair, and content backlog.
Agency Client
Before/after evidence, competitor movement, and monthly action plan.
Source Repair
Owned citation loss, third-party trust gaps, source freshness, and the pages to update first.
CLI verification
Run npx prompts-gpt orchestrate --mode eval --dry-run --eval-criteria correctness,risk from a synced workspace to verify the orchestration path before executing real agents.