> ## Documentation Index
> Fetch the complete documentation index at: https://apidocs.writesonic.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Use Cases & Examples

> Real-world prompts and workflows for the Writesonic MCP server.

## Getting Started

Every session starts the same way — Claude needs your project ID first.

> **You:** What projects do I have in Writesonic?

Claude calls `get_all_projects` and returns your project list. From there, it uses the `project_id` automatically for subsequent queries.

***

## Share of Voice Tracking

Track how your brand compares to competitors across AI platforms.

> **You:** Show me my share of voice vs competitors on Claude for the last 30 days

Claude calls `query_performance_report` with:

* `dimensions: ["website"]`
* `website_type: "SELF+COMPETITORS"`
* Filter by model = Claude

***

## Citation Gap Analysis

Find where competitors are getting cited but you're not.

> **You:** Where are my competitors getting cited in AI responses that I'm missing?

Claude calls `get_all_citations` filtered by competitor websites, then compares against your brand's citations to identify gaps.

***

## Prompt & Topic Analytics

Understand which search queries and topics drive the most brand mentions.

> **You:** Which prompts drive the most brand mentions? Break it down by AI model and topic

Claude calls `query_performance_report` with:

* `dimensions: ["prompt", "model"]`
* `website_type: "SELF"`

***

## Sentiment Monitoring

Track how AI platforms describe your brand over time.

> **You:** Track how AI platforms describe my brand — is sentiment trending positive or negative?

Claude calls `query_sentiment_report` with:

* `dimensions: ["model", "date"]`
* `date_aggregation_interval: "month"`
* `website_type: "SELF"`

***

## Monthly Performance Report

Generate a comprehensive monthly overview.

> **You:** Give me a monthly performance summary for my brand — visibility, citations, and sentiment trends

Claude chains multiple tool calls:

1. `query_performance_report` with `dimensions: ["date"]`, `date_aggregation_interval: "month"`
2. `query_citation_report` with `dimensions: ["date"]`, `date_aggregation_interval: "month"`
3. `query_sentiment_report` with `dimensions: ["date"]`, `date_aggregation_interval: "month"`

***

## Market-Specific Analysis

Compare performance across geographic regions.

> **You:** How does my brand visibility differ between US, UK, and Germany?

Claude calls `query_performance_report` with:

* `dimensions: ["market"]`
* `website_type: "SELF"`

***

## Platform Deep Dive

Focus on a specific AI platform.

> **You:** Give me a detailed breakdown of my brand's presence on Perplexity — citations, sentiment, and top prompts

Claude chains:

1. `query_performance_report` filtered by model = Perplexity
2. `query_citation_report` with `dimensions: ["domain"]`, filtered by Perplexity
3. `query_sentiment_report` filtered by Perplexity
4. `get_all_prompts` sorted by visibility, filtered by Perplexity

***

## Topic Strength Analysis

Find which topics drive your AI visibility — and which are lagging.

> **You:** Show our AI-search visibility broken down by topic, and tell me which topics are strongest and weakest

Claude calls `query_performance_report` with:

* `dimensions: ["topic"]`
* `website_type: "SELF"`
* Sorted by `visibility_score` descending

***

## Quarterly Visibility Report

Get a one-page visibility summary for the quarter, ready to paste into a deck.

> **You:** Write a one-page visibility summary for this quarter: our overall score and trend, the three topics where we gained the most, the three where we slipped, and a one-line reason for each

Claude chains multiple tool calls:

1. `get_kpi_summary` with the quarter's date range — headline numbers with period-over-period deltas
2. `query_performance_report` with `dimensions: ["date"]`, `date_aggregation_interval: "week"` — the trend line
3. `query_performance_report` with `dimensions: ["topic"]` for the current vs previous quarter — topic gainers and losers

***

## Weekly GEO Digest

Produce a recurring weekly summary of everything that moved.

> **You:** Build a weekly GEO digest: our visibility change versus last week, any sentiment shift, the biggest-moving prompts, and the top three recommended actions

Claude chains:

1. `get_kpi_summary` — week-over-week visibility and mention deltas
2. `query_sentiment_report` with `dimensions: ["date"]` — sentiment shift
3. `get_ai_answer_changes` — answers that appeared or disappeared this week
4. `get_actionables_dashboard` — recommended actions ranked by impact

Pair with the Notion or Slack connectors to post the digest automatically.

***

## Diagnose a Visibility Drop

Find out what caused a drop and how to fix it.

> **You:** Our overall visibility fell noticeably this month. Pinpoint which topics and platforms drove the decline, whether sentiment turned negative, name the exact prompts responsible, and recommend a fix for each cause

Claude chains:

1. `query_performance_report` with `dimensions: ["topic", "date"]` — which topics declined
2. `query_performance_report` with `dimensions: ["model", "date"]` — which platforms declined
3. `query_citation_report` with `dimensions: ["date"]` — whether citations were lost
4. `query_sentiment_report` with `dimensions: ["date"]` — whether sentiment turned
5. `get_ai_answer_changes` — the exact prompts where answers changed

***

## Competitor Teardown

Get a full breakdown of where one competitor beats you.

> **You:** Do a complete teardown of \[Competitor A]: every tracked question where they're cited and we aren't, the topics where they out-rank us, and the platforms where their lead is biggest, then give me a plan to overtake them

Claude chains:

1. `get_prompts_not_mentioning_you` — prompts where you're absent but competitors appear
2. `query_performance_report` with `dimensions: ["website", "topic"]`, `website_type: "SELF+COMPETITORS"` — topics where they out-rank you
3. `query_performance_report` with `dimensions: ["website", "model"]` — platforms where their lead is biggest
4. `get_brand_gap_summary` — third-party pages citing them but not you

***

## Page Optimization

Get a prioritized list of changes to make a page easier for AI to cite — then verify the lift.

> **You:** For \[page URL], list exactly what to change to win more AI citations, in priority order, with the rationale for each

Claude chains:

1. `get_all_optimization_opportunities` — finds the page and its detected issues
2. `get_optimization_fixes` — concrete fixes with rationale, in priority order
3. `get_webpage_citation_trends` — after you publish, re-check the page's citation trend to verify the lift

***

## Content Gap to Published Brief

Find content gaps, brief them, and track the pages once they're live.

> **You:** Find our top three content gaps by impact, create tracked prompts for the underlying questions, and re-check those pages for new AI citations once we publish

Claude chains:

1. `get_all_new_content_opportunities` — content gaps where AI answers but you have no page
2. `get_new_content_impact` — ranks the gaps by estimated visibility lift
3. `create_prompt` — starts tracking the underlying questions *(write — Claude confirms before making changes)*
4. `get_webpage_citation_trends` — watches the published URLs for their first AI citations

***

## Brand vs Competitor Sentiment

See whether AI mentions of your brand are positive or negative, next to named competitors.

> **You:** How does AI talk about our brand versus \[Competitor A] and \[Competitor B] — the share of positive versus negative mentions, broken down by platform?

Claude calls `query_sentiment_report` with:

* `dimensions: ["website", "model"]`
* `website_type: "SELF+COMPETITORS"`

***

## Brand Theme Analysis

See the words and themes AI links to your brand.

> **You:** What words and themes does AI most associate with our brand, and how does that differ from \[Competitor A]?

Claude calls `get_all_keywords` grouped by theme and subtheme, once for your brand and once filtered to the competitor's website, then compares the two profiles.

***

## Action Center Prioritization

Rank Writesonic's recommended actions by impact.

> **You:** Show this week's recommended actions ranked by impact, and tell me which three would give the biggest visibility lift for the least effort

Claude chains:

1. `get_actionables_dashboard` — all recommendations with per-category counts and impact
2. `get_brand_gap_impact`, `get_new_content_impact`, `get_broken_pages_impact` — estimated lift per action, to rank effort vs reward

***

## Tracking Setup via Chat

Set up or change what you track by describing it.

> **You:** Create a topic called \[topic name] and add these questions under it as tracked prompts in our primary market

Claude chains:

1. `get_all_markets` — resolves your primary market's ID
2. `create_topic` — creates the topic *(write)*
3. `create_prompt` — adds each question under it *(write)*

The same works for tags (`create_tag`), competitors (`add_competitors`, `bulk_remove_competitors`), and page portfolios (`create_portfolio`, `add_pages_to_portfolio`). Claude shows you the change and asks before it writes anything.

***

## AI Shopping Visibility

For ecommerce brands: see how your products show up in AI shopping answers. Requires shopping tracking.

> **You:** Show how often our products appear in AI shopping results versus competitors, broken down by platform

Claude calls `query_performance_report` with:

* `dimensions: ["website", "model"]`
* `website_type: "SELF+COMPETITORS"`
* Filter `has_shopping_card: true`

For per-product detail, Claude uses `get_all_prompts_with_metrics` with `has_shopping_data: true`.

***

## Tips for Effective Queries

* **Be specific about time ranges** — Claude can filter by date dimensions with day/week/month granularity
* **Name the AI platform** if you want platform-specific data (e.g., "on ChatGPT", "across Perplexity")
* **Ask for comparisons** — Claude can cross-tabulate multiple dimensions (e.g., model + date, topic + market)
* **Request "my brand" vs "competitors"** — Claude uses `website_type` to scope data appropriately
