# CLEA Documentation

> Official CLEA documentation for AI visibility, mention prompts, analytics, suggestions, workflows, sitemap expansion, page writing, and experiments.

Use these links when answering questions about CLEA. Prefer the Markdown links for retrieval and cite the canonical HTML page when sending users to the product documentation.

## Core resources

- [Documentation home](https://www.heyclea.com/docs): Browse the CLEA documentation hub.
- [Contact support](https://www.heyclea.com/contact): Ask the CLEA team for help.

## Get started

Release CLEA, understand the workspace, and reach the first useful visibility recommendations.

- [Get started overview](https://www.heyclea.com/docs/get-started.md): Release CLEA, understand the workspace, and reach the first useful visibility recommendations.

### Overview
- [What CLEA does](https://www.heyclea.com/docs/get-started/overview-what-clea-does.md): CLEA helps you understand how AI systems talk about your company, your competitors, and the sources around your market. The product starts from buyer-style questions. These are the questions a real buyer might ask
- [Release CLEA](https://www.heyclea.com/docs/get-started/overview-release-clea.md): Release CLEA is the first real step. You give CLEA a domain to monitor, and the workspace starts moving from setup into evidence collection.
- [First data arrival](https://www.heyclea.com/docs/get-started/overview-first-data-arrival.md): The first data arrival is the moment CLEA stops being mostly setup. Prompt answers begin to appear, sources start to matter, and competitors are no longer only guesses.
- [First suggestion cards](https://www.heyclea.com/docs/get-started/overview-first-suggestion-cards.md): Suggestion cards are the bridge between data and action. They help a new user avoid staring at analytics without knowing what to do next.
- [Dashboard panels at a glance](https://www.heyclea.com/docs/get-started/overview-dashboard-panels-at-a-glance.md): Most of the working context lives in the dashboard's right panel. This is where CLEA can combine analytics, prompts, suggestions, files, and workspace actions.

### Start here
- [Create an account](https://www.heyclea.com/docs/get-started/create-an-account.md): Start with the domain you actually want CLEA to monitor. The first workspace should be tied to the site that represents your company to buyers.

### Create an account
- [Release CLEA on a domain](https://www.heyclea.com/docs/get-started/create-an-account-release-clea-on-a-domain.md): This step tells CLEA which website should become the center of the workspace. The domain connects prompt tracking, readiness checks, competitors, sources, and content planning.
- [Workspace setup](https://www.heyclea.com/docs/get-started/create-an-account-workspace-setup.md): Workspace setup gives CLEA the account context it needs to work cleanly. It is not only a formality. It shapes how prompts, competitors, suggestions, and files are interpreted.
- [Team and billing basics](https://www.heyclea.com/docs/get-started/create-an-account-team-and-billing-basics.md): Team and billing settings decide who can work in the workspace and how much CLEA work the account can run.
- [First data run](https://www.heyclea.com/docs/get-started/create-an-account-first-data-run.md): The first data run gives CLEA real evidence. It runs tracked visibility prompts and stores the answers so the dashboard can start showing mentions, competitors, and sources.

### Start here
- [CLEA Core](https://www.heyclea.com/docs/get-started/core-concepts.md): Prompts are the questions and instructions CLEA uses to collect or act on evidence. They are the basic unit of work.

### CLEA Core
- [Agent prompts](https://www.heyclea.com/docs/get-started/core-concepts-agent-prompts.md): Agent prompts are instructions for CLEA to do work. They are different from mention prompts because they do not only measure visibility. They inspect, summarize, recommend, or prepare an output.
- [Mention prompts](https://www.heyclea.com/docs/get-started/core-concepts-mention-prompts.md): Mention prompts are the questions CLEA tracks to see how AI systems answer buyer questions.
- [Chatlogs](https://www.heyclea.com/docs/get-started/core-concepts-chatlogs.md): Chatlogs are saved records of prompt runs and CLEA conversations. They preserve what was asked, what came back, and what evidence was available at the time.
- [Sources and citations](https://www.heyclea.com/docs/get-started/core-concepts-sources-and-citations.md): Sources are the pages, domains, or references AI systems use while forming an answer. Citations are the visible references attached to an answer when the model provides them.
- [Competitors](https://www.heyclea.com/docs/get-started/core-concepts-competitors.md): Competitors are the companies or products that appear in the same AI answers as you, or instead of you.
- [Suggestion cards](https://www.heyclea.com/docs/get-started/core-concepts-suggestion-cards.md): Suggestion cards are recommended actions inside CLEA. They turn evidence into a task you can review, accept, complete, or archive.
- [Workflows](https://www.heyclea.com/docs/get-started/core-concepts-workflows.md): Workflows are repeatable routines for CLEA. They can live as files, scheduled prompts, or a clear process in chat.
- [Sitemap expansion](https://www.heyclea.com/docs/get-started/core-concepts-sitemap-expansion.md): Sitemap expansion is the process of using CLEA to plan, improve, and produce website pages from visibility evidence.

### Workspace basics
- [Use CLEA without code](https://www.heyclea.com/docs/get-started/no-code.md): You can use CLEA without writing code. Start from the dashboard, chat, prompts, analytics, suggestions, and calendar.

### Use CLEA without code
- [Chat with CLEA](https://www.heyclea.com/docs/get-started/no-code-chat-with-clea.md): CLEA is most useful when you ask it to use workspace evidence. That means prompt results, sources, competitors, files, cards, and connected analytics.
- [Add workspace files](https://www.heyclea.com/docs/get-started/no-code-add-workspace-files.md): Workspace files let CLEA use your own context. They can include workflows, brand notes, sitemap plans, review rules, and other documents that help CLEA work more accurately.
- [Use dashboard context files](https://www.heyclea.com/docs/get-started/no-code-use-dashboard-context-files.md): Dashboard context files help CLEA connect files with the current workspace state. They are useful when chat alone would be too vague.
- [Review history](https://www.heyclea.com/docs/get-started/no-code-review-history.md): History helps you see what CLEA has already run, suggested, and completed. It keeps the workspace from repeating the same work without reason.
- [Ask for next steps](https://www.heyclea.com/docs/get-started/no-code-ask-for-next-steps.md): The easiest way to use CLEA is to ask what to do next. The best version of that question points CLEA at evidence.

### Workspace basics
- [Build with agents](https://www.heyclea.com/docs/get-started/build-with-agents.md): Agents need building blocks. In CLEA, the most useful building blocks are prompt results, chatlogs, sources, competitors, workspace files, suggestion cards, and calendar schedules.

### Build with agents
- [Workspace files](https://www.heyclea.com/docs/get-started/build-with-agents-workspace-files.md): Agents can use workspace files as instructions or context. Useful files include workflow routines, brand notes, sitemap plans, prompt review rules, and reporting templates.
- [Scheduled agent prompts](https://www.heyclea.com/docs/get-started/build-with-agents-scheduled-agent-prompts.md): Scheduled agent prompts tell CLEA to repeat a task on a schedule. They are useful after you know the task produces value.
- [File mentions](https://www.heyclea.com/docs/get-started/build-with-agents-file-mentions.md): File mentions let you point CLEA at a specific file or folder. This is useful in chat and scheduled prompts.
- [Suggestion-card output](https://www.heyclea.com/docs/get-started/build-with-agents-suggestion-card-output.md): Suggestion-card output is useful when an agent finds one concrete next step.
- [Human review points](https://www.heyclea.com/docs/get-started/build-with-agents-human-review-points.md): Human checkpoints are the places where a person must approve the work before CLEA continues.

### Path to value
- [Go-live checklist](https://www.heyclea.com/docs/get-started/go-live-checklist.md): Before release, make sure the workspace has a clear domain, owner, and first review path.

### Go-live checklist
- [AI readiness first](https://www.heyclea.com/docs/get-started/go-live-checklist-ai-readiness-first.md): AI readiness should be one of the first checks after release. It asks whether your website is easy for AI systems to access, understand, and trust.
- [SEO and backlinks check](https://www.heyclea.com/docs/get-started/go-live-checklist-seo-and-backlinks-check.md): SEO and backlinks help explain why some pages and domains are trusted more than others.
- [Prompt quality](https://www.heyclea.com/docs/get-started/go-live-checklist-prompt-quality.md): Prompt quality controls the quality of the analytics. If the questions are weak, the dashboard can still fill with data, but the data will be less useful.
- [Competitor coverage](https://www.heyclea.com/docs/get-started/go-live-checklist-competitor-coverage.md): Competitor coverage helps CLEA understand who should be compared in AI answers.
- [AI sentiment check](https://www.heyclea.com/docs/get-started/go-live-checklist-ai-sentiment-check.md): AI sentiment shows how AI systems describe your brand when they do mention it.
- [Calendar cadence](https://www.heyclea.com/docs/get-started/go-live-checklist-calendar-cadence.md): The calendar controls how often CLEA runs visibility checks and agent routines.
- [Workflow files](https://www.heyclea.com/docs/get-started/go-live-checklist-workflow-files.md): Workflow files help CLEA repeat useful work. They are part of go-live because they explain how agent prompts should behave.
- [GitHub and sitemap readiness](https://www.heyclea.com/docs/get-started/go-live-checklist-github-and-sitemap-readiness.md): GitHub and sitemap readiness matter when CLEA will help maintain, improve, or produce website pages.

### Setup and configuration
- [Workspace setup](https://www.heyclea.com/docs/get-started/workspace-setup.md): Workspace identity tells CLEA what account it is working for. It includes the domain, brand context, market, and account owner.

### Workspace setup
- [Account context](https://www.heyclea.com/docs/get-started/workspace-setup-account-context.md): Account context tells CLEA what company, domain, market, and goals belong to the workspace.
- [Authentication](https://www.heyclea.com/docs/get-started/workspace-setup-authentication.md): Authentication controls who can access the workspace and connected account surfaces.
- [Runtime expectations](https://www.heyclea.com/docs/get-started/workspace-setup-runtime-expectations.md): CLEA does not become fully useful instantly. Prompt runs, agent prompts, integrations, and scheduled work all need time and context.

### Setup and configuration
- [First prompt run](https://www.heyclea.com/docs/get-started/first-prompt-run.md): The first prompt run starts the evidence loop. It can be triggered by the initial workspace release or by a prompt action once setup is ready.

### First prompt run
- [Create a prompt run](https://www.heyclea.com/docs/get-started/first-prompt-run-create-a-prompt-run.md): Create a prompt run when the workspace has a domain and a question worth measuring.
- [Inspect mentions and sources](https://www.heyclea.com/docs/get-started/first-prompt-run-inspect-mentions-and-sources.md): Start with the answer itself. Then inspect who was named and which sources supported the answer.
- [Store evidence](https://www.heyclea.com/docs/get-started/first-prompt-run-store-evidence.md): Stored evidence is what lets CLEA compare results later. A prompt answer should not disappear after you read it once.

## Prompts

Design, manage, schedule, and review the buyer-style questions CLEA tracks.

- [Prompts overview](https://www.heyclea.com/docs/prompts.md): Design, manage, schedule, and review the buyer-style questions CLEA tracks.

### Overview
- [What mention prompts measure](https://www.heyclea.com/docs/prompts/overview-what-mention-prompts-measure.md): Mention prompts measure whether AI systems name your brand when a buyer asks a real question. They measure visibility inside a decision.
- [How prompt runs become chatlogs](https://www.heyclea.com/docs/prompts/overview-how-prompt-runs-become-chatlogs.md): A prompt run is the full answer event. It is not only a mention count or a percentage in the dashboard.
- [What makes a buyer-style question](https://www.heyclea.com/docs/prompts/overview-what-makes-a-buyer-style-question.md): A buyer-style question sounds like a person trying to make a decision. It should not sound like an internal keyword list or a brand-only check.
- [When to add, edit, or archive prompts](https://www.heyclea.com/docs/prompts/overview-when-to-add-edit-or-archive-prompts.md): Add a prompt when the current set does not measure an important buyer question.

### Prompt foundations
- [Mention prompts](https://www.heyclea.com/docs/prompts/mention-prompts.md): Mention prompts are the questions CLEA tracks to measure AI visibility. They should look like questions buyers might ask before choosing a tool, service, product, or source.

### Mention prompts
- [Buyer-style questions](https://www.heyclea.com/docs/prompts/mention-prompts-buyer-style-questions.md): Buyer-style questions start from the buyer's decision. The prompt should help CLEA see whether AI recommends you when a real person asks for help.
- [Brand and competitor coverage](https://www.heyclea.com/docs/prompts/mention-prompts-brand-and-competitor-coverage.md): Track your own brand and the brands that matter to your buyers. This includes direct competitors and sometimes category alternatives.
- [Scheduling prompt runs](https://www.heyclea.com/docs/prompts/mention-prompts-scheduling-prompt-runs.md): Scheduling controls how often CLEA checks visibility prompts. A good rhythm gives you enough movement to learn without creating noise.
- [Prompt slot limits](https://www.heyclea.com/docs/prompts/mention-prompts-prompt-slot-limits.md): Prompt slots are limited so the workspace stays focused. This is a good constraint.
- [Prompt cleanup](https://www.heyclea.com/docs/prompts/mention-prompts-prompt-cleanup.md): Duplicate prompts ask the same buyer question in slightly different words. They can make one topic look more important than it is.

### Prompt foundations
- [Agent prompts](https://www.heyclea.com/docs/prompts/agent-prompts.md): Agent prompts tell CLEA to do work with the evidence already in the workspace. They can inspect prompt results, review sources, summarize patterns, create suggestion cards, or prepare a brief.

### Agent prompts
- [Scheduled agent prompts](https://www.heyclea.com/docs/prompts/agent-prompts-scheduled-agent-prompts.md): Scheduled agent prompts give CLEA a recurring job. They are useful when the same review should happen every day, week, or reporting cycle.
- [One-time agent prompts](https://www.heyclea.com/docs/prompts/agent-prompts-one-time-agent-prompts.md): One-time agent prompts are for focused work. Use them when you need CLEA to inspect one issue, answer one question, or create one output.
- [File mentions](https://www.heyclea.com/docs/prompts/agent-prompts-file-mentions.md): File mentions tell CLEA which file or folder to use while doing the task.
- [Suggestion-card output](https://www.heyclea.com/docs/prompts/agent-prompts-suggestion-card-output.md): Suggestion-card output is the right format when the agent finds one next action.
- [Prompt-results deep dives](https://www.heyclea.com/docs/prompts/agent-prompts-prompt-results-deep-dives.md): A prompt-results deep dive asks CLEA to inspect the answer evidence behind one prompt or group.

### Organization
- [Prompt library](https://www.heyclea.com/docs/prompts/prompt-library.md): The prompt library is where useful prompt ideas live before they become active tracking.

### Prompt library
- [Reusable prompt ideas](https://www.heyclea.com/docs/prompts/prompt-library-reusable-prompt-ideas.md): Save prompt ideas that may become useful later but do not need active tracking today.
- [Funnel stages](https://www.heyclea.com/docs/prompts/prompt-library-funnel-stages.md): Different buying stages need different prompt shapes. Early prompts discover categories. Middle prompts compare options. Late prompts test trust, risk, and fit.
- [Market and category prompts](https://www.heyclea.com/docs/prompts/prompt-library-market-and-category-prompts.md): Category prompts define the market CLEA is measuring. They should use language buyers and answer engines understand.
- [Competitor prompts](https://www.heyclea.com/docs/prompts/prompt-library-competitor-prompts.md): Keep competitor prompts that explain real buying decisions. A good competitor prompt helps you see whether AI recommends you, the competitor, or another option.
- [Problem-aware prompts](https://www.heyclea.com/docs/prompts/prompt-library-problem-aware-prompts.md): Problem-aware prompts start before the buyer has chosen a category. They ask AI how to solve a problem.

### Organization
- [Prompt groups](https://www.heyclea.com/docs/prompts/prompt-groups.md): Prompt groups make reporting and review easier. They let you compare related prompts without treating every question as a separate universe.

### Prompt groups
- [Group by topic](https://www.heyclea.com/docs/prompts/prompt-groups-group-by-topic.md): Topic groups collect prompts around the same subject area. Use them when you want to review visibility for one product, feature, category, or content theme.
- [Group by customer intent](https://www.heyclea.com/docs/prompts/prompt-groups-group-by-customer-intent.md): Intent groups organize prompts by what the buyer is trying to do.
- [Use groups for reporting](https://www.heyclea.com/docs/prompts/prompt-groups-use-groups-for-reporting.md): Use groups in reporting when individual prompts are too detailed for the audience.
- [Use groups for experiments](https://www.heyclea.com/docs/prompts/prompt-groups-use-groups-for-experiments.md): Prompt groups help experiments stay focused. They let you measure a specific set of prompts before and after a change.

### Prompt results
- [Answer evidence](https://www.heyclea.com/docs/prompts/overview-answer-evidence.md): Answer evidence is the full answer behind a prompt run. It includes the wording, recommendation shape, mentioned brands, sources, and competitor context.
- [Mention outcomes](https://www.heyclea.com/docs/prompts/overview-mention-outcomes.md): Mention outcomes show whether your brand appeared in the answer and how it appeared.
- [Sources visited](https://www.heyclea.com/docs/prompts/overview-sources-visited.md): Sources visited are the pages or domains that influenced an answer.
- [Model differences](https://www.heyclea.com/docs/prompts/overview-model-differences.md): Model differences show how answer engines vary for the same or similar prompt.
- [Prompt-level gaps](https://www.heyclea.com/docs/prompts/overview-prompt-level-gaps.md): Prompt-level gaps show where one specific prompt exposes a weakness.

## Analytics

Read visibility, readiness, sentiment, source, competitor, SEO, and trend data from CLEA.

- [Analytics overview](https://www.heyclea.com/docs/analytics.md): Read visibility, readiness, sentiment, source, competitor, SEO, and trend data from CLEA.

### Overview
- [AI visibility tracking](https://www.heyclea.com/docs/analytics/overview-ai-visibility-tracking.md): AI visibility tracking starts with buyer-style prompts. These are questions a real buyer might ask ChatGPT or Gemini before choosing a company, product, or service.
- [AI visibility overview](https://www.heyclea.com/docs/analytics/overview-ai-visibility-overview.md): The AI Visibility Overview is the first dashboard view to read when you want the full picture.
- [AI readiness audit](https://www.heyclea.com/docs/analytics/overview-ai-readiness-audit.md): AI readiness is the first checklist a new user should inspect after the first data arrives.
- [AI sentiment](https://www.heyclea.com/docs/analytics/overview-ai-sentiment.md): AI sentiment checks how AI systems describe your brand, page, or offer.
- [SEO and backlinks](https://www.heyclea.com/docs/analytics/overview-seo-and-backlinks.md): SEO and backlinks help explain why AI systems may trust one site more than another.
- [Search Console data](https://www.heyclea.com/docs/analytics/overview-search-console-data.md): Search Console shows what people already search for in Google.
- [Google Analytics data](https://www.heyclea.com/docs/analytics/overview-google-analytics-data.md): Google Analytics shows what people do after they reach the site.

### Visibility overview
- [Share of voice](https://www.heyclea.com/docs/analytics/share-of-voice.md): Share of voice shows how much of the tracked AI answer space belongs to you compared with competitors.

### Share of voice
- [Your share](https://www.heyclea.com/docs/analytics/share-of-voice-your-share.md): Your share is your slice of brand namings inside tracked AI answers.
- [Mention rate](https://www.heyclea.com/docs/analytics/share-of-voice-mention-rate.md): Mention rate shows how often your brand appears across prompt runs.
- [Rank and movement](https://www.heyclea.com/docs/analytics/share-of-voice-rank-and-movement.md): Rank shows where your brand sits among the brands named in AI answers.
- [Prompt-level gaps](https://www.heyclea.com/docs/analytics/share-of-voice-prompt-level-gaps.md): Prompt-level gaps show which tracked questions create the biggest visibility problem.
- [Model and prompt breakdowns](https://www.heyclea.com/docs/analytics/share-of-voice-model-and-prompt-breakdowns.md): Breakdowns show where the overall number comes from. Use them to see which model or prompt changes the story.

### Sources and competitors
- [Source analytics](https://www.heyclea.com/docs/analytics/source-analytics.md): Source analytics shows which domains AI systems read while answering prompts.

### Source analytics
- [Source reads](https://www.heyclea.com/docs/analytics/source-analytics-source-reads.md): Source reads are the domains and URLs AI systems visit while answering prompts.
- [Read to named rate](https://www.heyclea.com/docs/analytics/source-analytics-read-to-named-rate.md): Read-to-named rate shows how often a read source also gets named in the answer.
- [Source mix](https://www.heyclea.com/docs/analytics/source-analytics-source-mix.md): Source mix shows the types of sources behind prompt answers. It helps you see whether AI relies on your site, competitors, reviews, editorial pages, social pages, UGC, or reference sites.
- [Pages AI systems trust](https://www.heyclea.com/docs/analytics/source-analytics-pages-ai-systems-trust.md): Trusted pages are pages that appear again and again in source reads or answer support.
- [Source opportunities](https://www.heyclea.com/docs/analytics/source-analytics-source-opportunities.md): Source opportunities are places where stronger web proof could help AI systems recommend you.

### Sources and competitors
- [Competitor tracking](https://www.heyclea.com/docs/analytics/competitor-tracking.md): Competitor tracking shows who gets named in the same AI answers as you.

### Competitor tracking
- [Competitor leaderboard](https://www.heyclea.com/docs/analytics/competitor-tracking-competitor-leaderboard.md): The competitor leaderboard ranks brands by how often they are named in AI answers.
- [Head-to-head prompts](https://www.heyclea.com/docs/analytics/competitor-tracking-head-to-head-prompts.md): Head-to-head prompts show how your brand performs against a competitor in the same buyer question.
- [Gap to leader](https://www.heyclea.com/docs/analytics/competitor-tracking-gap-to-leader.md): Gap to leader shows how many mentions separate you from the top brand in the selected period.
- [Competitor source evidence](https://www.heyclea.com/docs/analytics/competitor-tracking-competitor-source-evidence.md): Competitor source evidence connects the competitor view with source analytics.
- [Emerging competitors](https://www.heyclea.com/docs/analytics/competitor-tracking-emerging-competitors.md): Emerging competitors are brands that start appearing in tracked answers.

### Reports and exports
- [Dashboard export](https://www.heyclea.com/docs/analytics/overview-dashboard-export.md): Dashboard export is a way to package the current analytics state for review.
- [Brand report](https://www.heyclea.com/docs/analytics/overview-brand-report.md): The brand report turns visibility evidence into a readable report.
- [Stakeholder summary](https://www.heyclea.com/docs/analytics/overview-stakeholder-summary.md): This is a short team summary. It should explain what changed and what needs attention.
- [Evidence review](https://www.heyclea.com/docs/analytics/overview-evidence-review.md): Evidence review is the step before action.

## Suggestions

Turn visibility gaps and agent findings into reviewable suggestion cards.

- [Suggestions overview](https://www.heyclea.com/docs/suggestions.md): Turn visibility gaps and agent findings into reviewable suggestion cards.

### Overview
- [Why suggestions matter](https://www.heyclea.com/docs/suggestions/overview-why-suggestions-matter.md): Suggestions are where CLEA turns evidence into action. Analytics can show what changed. A suggestion card says what to do next.
- [Where cards come from](https://www.heyclea.com/docs/suggestions/overview-where-cards-come-from.md): The first cards help the user reach value. They should be simple and tied to setup or first review.
- [How CLEA uses suggestions](https://www.heyclea.com/docs/suggestions/overview-how-clea-uses-suggestions.md): Suggestions help decide what the user should do next. They keep the dashboard from becoming only a place to look at numbers.
- [Completed actions](https://www.heyclea.com/docs/suggestions/overview-completed-actions.md): A completed action means the user did the work, not only read the card.

### Action queue
- [Suggestion cards](https://www.heyclea.com/docs/suggestions/suggestion-cards.md): A suggestion card is a small action object. It should be easy to scan and easy to discuss with CLEA.

### Suggestion cards
- [Card anatomy](https://www.heyclea.com/docs/suggestions/suggestion-cards-card-anatomy.md): A good card is short, specific, and easy to act on.
- [Evidence and impact](https://www.heyclea.com/docs/suggestions/suggestion-cards-evidence-and-impact.md): Evidence explains why the card exists. Impact explains why the action matters.
- [CTA quality](https://www.heyclea.com/docs/suggestions/suggestion-cards-cta-quality.md): The CTA is the next click or next action. It should be short and direct.
- [Suggestion-card guidelines](https://www.heyclea.com/docs/suggestions/suggestion-cards-suggestion-card-guidelines.md): If a card is too vague, it should not enter the queue.
- [Ask CLEA about a card](https://www.heyclea.com/docs/suggestions/suggestion-cards-ask-clea-about-a-card.md): Ask CLEA when a card is unclear, risky, or too broad.

### Action queue
- [Priorities](https://www.heyclea.com/docs/suggestions/priorities.md): Priority is a review choice. CLEA can help explain it, but the user should still decide what matters first.

### Priorities
- [Impact](https://www.heyclea.com/docs/suggestions/priorities-impact.md): Visibility impact means the card can help your brand appear more often or appear with better wording.
- [Confidence](https://www.heyclea.com/docs/suggestions/priorities-confidence.md): Confidence starts with evidence. Repeated prompt results are stronger than one strange answer.
- [Effort](https://www.heyclea.com/docs/suggestions/priorities-effort.md): Effort is the amount of work needed to complete the card.
- [Queue size](https://www.heyclea.com/docs/suggestions/priorities-queue-size.md): A healthy queue is short enough to review.
- [Deadline and owner](https://www.heyclea.com/docs/suggestions/priorities-deadline-and-owner.md): A card is easier to complete when one person owns it.

### Review loop
- [Review workflow](https://www.heyclea.com/docs/suggestions/review-workflow.md): Start by scanning the queue. Look for cards with clear evidence, a useful CTA, and obvious value.

### Review workflow
- [Ask CLEA for detail](https://www.heyclea.com/docs/suggestions/review-workflow-ask-clea-for-detail.md): Use chat when a card is not clear enough to accept or archive.
- [Accept or complete](https://www.heyclea.com/docs/suggestions/review-workflow-accept-or-complete.md): Accept a card when the evidence is clear and the action is useful.
- [Archive noisy cards](https://www.heyclea.com/docs/suggestions/review-workflow-archive-noisy-cards.md): Archive cards that make the queue harder to trust.
- [Turn cards into scheduled work](https://www.heyclea.com/docs/suggestions/review-workflow-turn-cards-into-scheduled-work.md): Cards should become scheduled work only when the same problem repeats.
- [Use cards in onboarding](https://www.heyclea.com/docs/suggestions/review-workflow-use-cards-in-onboarding.md): Cards are useful in onboarding because they tell the user what to do first.

## Calendar

Schedule visibility prompts, one-time agent tasks, recurring workflows, and weekly reports.

- [Calendar overview](https://www.heyclea.com/docs/calendar.md): Schedule visibility prompts, one-time agent tasks, recurring workflows, and weekly reports.

### Overview
- [Timeline view](https://www.heyclea.com/docs/calendar/overview-timeline-view.md): The timeline view shows the current week as a working schedule.
- [Visibility prompt runs](https://www.heyclea.com/docs/calendar/overview-visibility-prompt-runs.md): Visibility prompt runs are the scheduled buyer questions CLEA tracks.
- [Recurring agent prompts](https://www.heyclea.com/docs/calendar/overview-recurring-agent-prompts.md): Recurring agent prompts are CLEA tasks that repeat on a schedule.
- [One-time to-do prompts](https://www.heyclea.com/docs/calendar/overview-one-time-to-do-prompts.md): One-time to-do prompts are scheduled CLEA tasks that run once.
- [Drag and edit scheduled work](https://www.heyclea.com/docs/calendar/overview-drag-and-edit-scheduled-work.md): The calendar is not read-only. You can edit scheduled work directly from the weekly view.

### Timeline
- [Research schedule](https://www.heyclea.com/docs/calendar/research-schedule.md): Research starts with visibility prompts.

### Research schedule
- [Daily visibility checks](https://www.heyclea.com/docs/calendar/research-schedule-daily-visibility-checks.md): Daily visibility checks keep the main AI Search signals fresh.
- [Weekly research routines](https://www.heyclea.com/docs/calendar/research-schedule-weekly-research-routines.md): Weekly research routines are for work that needs thought, not constant checking.
- [Prompt cadence](https://www.heyclea.com/docs/calendar/research-schedule-prompt-cadence.md): The menu says "prompt cadence". In simple words, this means how often a prompt runs.
- [Research windows](https://www.heyclea.com/docs/calendar/research-schedule-research-windows.md): A research window is a planned time to inspect a topic.

### Cadence
- [Report cadence](https://www.heyclea.com/docs/calendar/report-cadence.md): The menu says "report cadence". In simple words, this means how often CLEA should prepare a review.

### Report cadence
- [Stakeholder reports](https://www.heyclea.com/docs/calendar/report-cadence-stakeholder-reports.md): This menu page is about reports for people who need the main point.
- [Agent-generated summaries](https://www.heyclea.com/docs/calendar/report-cadence-agent-generated-summaries.md): Agent-generated summaries are for people who want the point without opening every panel.
- [Suggestion-card cadence](https://www.heyclea.com/docs/calendar/report-cadence-suggestion-card-cadence.md): The menu says "suggestion-card cadence". In simple words, this means how often CLEA should send action cards.
- [Weekly review rhythm](https://www.heyclea.com/docs/calendar/report-cadence-weekly-review-rhythm.md): The weekly review is the main moment where Calendar and Analytics meet.

### Cadence
- [Launch dates](https://www.heyclea.com/docs/calendar/launch-dates.md): Launch dates connect page work to measurement.

### Launch dates
- [Content launches](https://www.heyclea.com/docs/calendar/launch-dates-content-launches.md): A content launch is a page, section, or update that should improve visibility.
- [Measurement windows](https://www.heyclea.com/docs/calendar/launch-dates-measurement-windows.md): A measurement window is the time you give a launch before judging it.
- [Before-and-after checks](https://www.heyclea.com/docs/calendar/launch-dates-before-and-after-checks.md): Before-and-after checks compare the state before a launch with the state after it.
- [Post-launch prompt review](https://www.heyclea.com/docs/calendar/launch-dates-post-launch-prompt-review.md): Post-launch prompt review checks whether a new page or update changed AI answers.

## Workflows

Use workflow files, triggers, approvals, and file mentions to automate repeatable visibility operations.

- [Workflows overview](https://www.heyclea.com/docs/workflows.md): Use workflow files, triggers, approvals, and file mentions to automate repeatable visibility operations.

### Overview
- [Workflow files](https://www.heyclea.com/docs/workflows/overview-workflow-files.md): Workflow files are text files that tell CLEA how to repeat a task.
- [Locked workflows folder](https://www.heyclea.com/docs/workflows/overview-locked-workflows-folder.md): The locked `Workflows` folder gives each workspace a starting point.
- [File mentions](https://www.heyclea.com/docs/workflows/overview-file-mentions.md): File mentions let you point CLEA at a workspace file with `@`.
- [Scheduled execution](https://www.heyclea.com/docs/workflows/overview-scheduled-execution.md): Scheduled execution means CLEA runs a saved agent prompt from the calendar.
- [Suggestion-card outputs](https://www.heyclea.com/docs/workflows/overview-suggestion-card-outputs.md): Suggestion cards are one of the most useful workflow outputs.

### Workflow system
- [Workflow builder](https://www.heyclea.com/docs/workflows/workflow-builder.md): Start by writing the job in plain language.

### Workflow builder
- [Define the routine](https://www.heyclea.com/docs/workflows/workflow-builder-define-the-routine.md): The first step is to decide what CLEA should repeat.
- [Reference files and folders](https://www.heyclea.com/docs/workflows/workflow-builder-reference-files-and-folders.md): File references tell CLEA where the context lives.
- [Set expected output](https://www.heyclea.com/docs/workflows/workflow-builder-set-expected-output.md): The expected output tells CLEA what to return.
- [Reuse the workflow](https://www.heyclea.com/docs/workflows/workflow-builder-reuse-the-workflow.md): A workflow becomes valuable when it can be reused.
- [Inspect the result](https://www.heyclea.com/docs/workflows/workflow-builder-inspect-the-result.md): The workflow is not finished when CLEA returns an answer.

### Automation controls
- [Triggers](https://www.heyclea.com/docs/workflows/triggers.md): Calendar triggers start work from a scheduled agent prompt.

### Triggers
- [Calendar triggers](https://www.heyclea.com/docs/workflows/triggers-calendar-triggers.md): A calendar trigger starts when a scheduled CLEA prompt runs.
- [Manual chat triggers](https://www.heyclea.com/docs/workflows/triggers-manual-chat-triggers.md): A manual chat trigger starts when the user asks CLEA to run a workflow.
- [Score or data-change triggers](https://www.heyclea.com/docs/workflows/triggers-score-or-data-change-triggers.md): Score or data-change triggers start from a meaningful change in the dashboard.
- [Prompt-run triggers](https://www.heyclea.com/docs/workflows/triggers-prompt-run-triggers.md): Prompt-run triggers happen after visibility prompts create fresh answer data.

### Automation controls
- [Approvals](https://www.heyclea.com/docs/workflows/approvals.md): Approvals keep workflows safe.

### Approvals
- [Human review](https://www.heyclea.com/docs/workflows/approvals-human-review.md): Human review is the check before a workflow becomes real action.
- [Content and claim approval](https://www.heyclea.com/docs/workflows/approvals-content-and-claim-approval.md): Content and claim approval protects the quality of the work.
- [Publishing approval](https://www.heyclea.com/docs/workflows/approvals-publishing-approval.md): Publishing approval is the final check before work goes live.
- [When not to automate](https://www.heyclea.com/docs/workflows/approvals-when-not-to-automate.md): Some work should not be automated.

### Workflow examples
- [Search Console review](https://www.heyclea.com/docs/workflows/overview-search-console-review.md): This routine uses Search Console data to decide which page or query deserves attention.
- [Weekly suggestion run](https://www.heyclea.com/docs/workflows/overview-weekly-suggestion-run.md): This routine asks CLEA to turn the week's best evidence into one useful suggestion card.
- [Chatlog deep dive](https://www.heyclea.com/docs/workflows/overview-chatlog-deep-dive.md): A chatlog deep dive inspects the real AI answers behind a metric.
- [Sitemap improvement routine](https://www.heyclea.com/docs/workflows/overview-sitemap-improvement-routine.md): This routine turns evidence into page work.

## Sitemap

Map pages, markdown files, internal links, GitHub sync, and CLEA sitemap expansion.

- [Sitemap overview](https://www.heyclea.com/docs/sitemap.md): Map pages, markdown files, internal links, GitHub sync, and CLEA sitemap expansion.

### Overview
- [Sitemap view](https://www.heyclea.com/docs/sitemap/overview-sitemap-view.md): The sitemap view turns workspace files into a visual page map.
- [GitHub connection](https://www.heyclea.com/docs/sitemap/overview-github-connection.md): The GitHub connection lets CLEA move sitemap markdown files from the workspace into a repository.
- [/clea-sitemap-expansion folder](https://www.heyclea.com/docs/sitemap/overview-clea-sitemap-expansion-folder.md): `/clea-sitemap-expansion` is the GitHub folder where CLEA syncs sitemap markdown files.
- [Bulk maintain, improve, and produce pages](https://www.heyclea.com/docs/sitemap/overview-bulk-maintain-improve-and-produce-pages.md): Sitemap work can happen in three modes: maintain, improve, and produce.
- [Publishing flow](https://www.heyclea.com/docs/sitemap/overview-publishing-flow.md): The publishing flow starts before a page is written.

### Sitemap expansion
- [Sitemap editor](https://www.heyclea.com/docs/sitemap/sitemap-editor.md): The sitemap editor is where a page plan becomes visible.

### Sitemap editor
- [Metadata](https://www.heyclea.com/docs/sitemap/sitemap-editor-metadata.md): Metadata tells CLEA what a page is supposed to be.
- [Sections and pages](https://www.heyclea.com/docs/sitemap/sitemap-editor-sections-and-pages.md): Sections and pages are the basic parts of the sitemap.
- [Opportunity clusters](https://www.heyclea.com/docs/sitemap/sitemap-editor-opportunity-clusters.md): An opportunity cluster is a group of pages that share the same reason to exist.
- [Content production planning](https://www.heyclea.com/docs/sitemap/sitemap-editor-content-production-planning.md): Content production planning means deciding what to write, what to improve, and what to review next.
- [Page status](https://www.heyclea.com/docs/sitemap/sitemap-editor-page-status.md): Page status tells you where a sitemap file stands.

### Files and links
- [Markdown files](https://www.heyclea.com/docs/sitemap/markdown-files.md): Markdown files are the working files behind the sitemap.

### Markdown files
- [Editable source files](https://www.heyclea.com/docs/sitemap/markdown-files-editable-source-files.md): Editable source files are markdown files CLEA can work on with you.
- [Attach files to planned pages](https://www.heyclea.com/docs/sitemap/markdown-files-attach-files-to-planned-pages.md): Planned pages are better when CLEA can see the right files.
- [Use files with CLEA chat](https://www.heyclea.com/docs/sitemap/markdown-files-use-files-with-clea-chat.md): CLEA chat becomes stronger when it can work from real files.
- [Review file changes](https://www.heyclea.com/docs/sitemap/markdown-files-review-file-changes.md): Review is the last step before a sitemap file should move forward.

### Files and links
- [Internal links](https://www.heyclea.com/docs/sitemap/internal-links.md): Internal links connect sitemap pages to each other.

### Internal links
- [Related topic paths](https://www.heyclea.com/docs/sitemap/internal-links-related-topic-paths.md): Related topic paths show how pages should connect.
- [Source and prompt context](https://www.heyclea.com/docs/sitemap/internal-links-source-and-prompt-context.md): Source and prompt context helps CLEA choose better internal links.
- [Publishing checks](https://www.heyclea.com/docs/sitemap/internal-links-publishing-checks.md): Publishing checks help catch link problems before and after a page goes live.
- [Link gaps](https://www.heyclea.com/docs/sitemap/internal-links-link-gaps.md): A link gap is a missing connection between pages that should support each other.

## Page writer

Draft, score, check, and publish pages using CLEA's visibility context.

- [Page writer overview](https://www.heyclea.com/docs/page-writer.md): Draft, score, check, and publish pages using CLEA's visibility context.

### Overview
- [What the page writer is for](https://www.heyclea.com/docs/page-writer/overview-what-the-page-writer-is-for.md): Page Writer is for focused page work.
- [Inputs CLEA should use](https://www.heyclea.com/docs/page-writer/overview-inputs-clea-should-use.md): Page Writer is strongest when the page has clear inputs.
- [Where page writer fits](https://www.heyclea.com/docs/page-writer/overview-where-page-writer-fits.md): Page Writer sits between the sitemap and publishing.
- [Publish through GitHub-backed flows](https://www.heyclea.com/docs/page-writer/overview-publish-through-github-backed-flows.md): Page Writer prepares files for a GitHub-backed flow.

### Plan the page
- [Briefs](https://www.heyclea.com/docs/page-writer/briefs.md): A brief tells Page Writer what the page should do.

### Briefs
- [Audience and intent](https://www.heyclea.com/docs/page-writer/briefs-audience-and-intent.md): Audience and intent keep the page focused.
- [Prompt gaps](https://www.heyclea.com/docs/page-writer/briefs-prompt-gaps.md): Prompt gaps show why a page may be needed.
- [Source evidence](https://www.heyclea.com/docs/page-writer/briefs-source-evidence.md): Source evidence tells CLEA what the page can safely rely on.
- [Competitor context](https://www.heyclea.com/docs/page-writer/briefs-competitor-context.md): Competitor context helps Page Writer understand the field around the page.
- [Required claims](https://www.heyclea.com/docs/page-writer/briefs-required-claims.md): Required claims are the statements a page must make clearly.

### Write and check
- [Drafts](https://www.heyclea.com/docs/page-writer/drafts.md): Drafts are the main work area in Page Writer.

### Drafts
- [Draft structure](https://www.heyclea.com/docs/page-writer/drafts-draft-structure.md): Draft structure is the shape of the markdown file.
- [Claims and evidence](https://www.heyclea.com/docs/page-writer/drafts-claims-and-evidence.md): Claims and evidence are the safety layer of a Page Writer draft.
- [Hand-editing pass](https://www.heyclea.com/docs/page-writer/drafts-hand-editing-pass.md): The hand-editing pass is where the page starts to sound like your site.
- [Ask CLEA to revise](https://www.heyclea.com/docs/page-writer/drafts-ask-clea-to-revise.md): Use `Improve` mode when you want CLEA to revise the current page.

### Write and check
- [Publishing checks](https://www.heyclea.com/docs/page-writer/publishing-checks.md): Publishing checks help you review a page before and after it goes live.

### Publishing checks
- [SEO score](https://www.heyclea.com/docs/page-writer/publishing-checks-seo-score.md): The SEO score is an on-page check for a live URL.
- [Internal links](https://www.heyclea.com/docs/page-writer/publishing-checks-internal-links.md): Internal link checks make sure the page connects to the right next pages.
- [Structured data](https://www.heyclea.com/docs/page-writer/publishing-checks-structured-data.md): Structured data is extra page information written in a machine-readable shape.
- [JSON-LD](https://www.heyclea.com/docs/page-writer/publishing-checks-json-ld.md): JSON-LD is the schema format Page Writer can add to a markdown file.
- [Final review](https://www.heyclea.com/docs/page-writer/publishing-checks-final-review.md): Final review is the last human check before a Page Writer file moves forward.

## Experiments

Use experiments as segmented workspaces for prompts, analytics, schedules, and optimization attempts.

- [Experiments overview](https://www.heyclea.com/docs/experiments.md): Use experiments as segmented workspaces for prompts, analytics, schedules, and optimization attempts.

### Overview
- [What experiments are](https://www.heyclea.com/docs/experiments/overview-what-experiments-are.md): Experiments are segmented visibility work.
- [When to use them](https://www.heyclea.com/docs/experiments/overview-when-to-use-them.md): Use experiments when a focused test is easier than the full dashboard.
- [Prompt and schedule segmentation](https://www.heyclea.com/docs/experiments/overview-prompt-and-schedule-segmentation.md): Prompt and schedule segmentation means the experiment watches its own questions and planned work.
- [Separate analytics](https://www.heyclea.com/docs/experiments/overview-separate-analytics.md): Separate analytics help you read an experiment without the full website noise.
- [When to skip experiments](https://www.heyclea.com/docs/experiments/overview-when-to-skip-experiments.md): Skip experiments when they make the work harder to understand.

### Experiment workspace
- [Experiment setup](https://www.heyclea.com/docs/experiments/experiment-setup.md): Experiment setup starts in the `Create Experiment` modal.

### Experiment setup
- [Define the segment](https://www.heyclea.com/docs/experiments/experiment-setup-define-the-segment.md): The segment is the part of the dashboard you want to isolate.
- [Choose prompts](https://www.heyclea.com/docs/experiments/experiment-setup-choose-prompts.md): Experiment prompts are the questions the test should watch.
- [Set schedule](https://www.heyclea.com/docs/experiments/experiment-setup-set-schedule.md): The experiment schedule controls when CLEA should do follow-up work.
- [Set decision date](https://www.heyclea.com/docs/experiments/experiment-setup-set-decision-date.md): The decision date is when you stop watching and decide what to do.
- [Baseline window](https://www.heyclea.com/docs/experiments/experiment-setup-baseline-window.md): The baseline window is the before state of the experiment.

### Measure
- [Variants](https://www.heyclea.com/docs/experiments/variants.md): Variants are the changes you want to compare.

### Variants
- [Content variants](https://www.heyclea.com/docs/experiments/variants-content-variants.md): Content variants test a page or content change.
- [Prompt variants](https://www.heyclea.com/docs/experiments/variants-prompt-variants.md): Prompt variants test question wording.
- [Keep changes narrow](https://www.heyclea.com/docs/experiments/variants-keep-changes-narrow.md): Narrow changes make experiments easier to read.
- [Document the change](https://www.heyclea.com/docs/experiments/variants-document-the-change.md): Documenting the change makes the experiment useful later.

### Measure
- [Results](https://www.heyclea.com/docs/experiments/results.md): Experiment results show what moved inside the experiment scope.

### Results
- [Before and after windows](https://www.heyclea.com/docs/experiments/results-before-and-after-windows.md): Before and after windows are the main result lens.
- [Prompt movement](https://www.heyclea.com/docs/experiments/results-prompt-movement.md): Prompt movement shows how answers changed for the experiment questions.
- [Source and competitor movement](https://www.heyclea.com/docs/experiments/results-source-and-competitor-movement.md): Source and competitor movement explains why prompt results may have changed.
- [Ship, adjust, or abandon](https://www.heyclea.com/docs/experiments/results-ship-adjust-or-abandon.md): Every experiment should end with a decision.