Integration

Writesonic integration

Writesonic tracks how a brand shows up in AI search answers (ChatGPT, Claude, Gemini, Google AI and others) and suggests content fixes. Export a site's tracked topics, prompts, AI platforms and competitor websites; read daily visibility, mention and rank metrics per website, per prompt or per raw AI answer, plus the pages AI cites and the keywords it uses; list content refresh pages with their issues, start AI fix generation for a page and read the suggested fixes; and read the latest SEO site audit issues. Every route is scoped to one site, passed as its URL. The older content generation routes (article writer, Chatsonic chat) are not exposed: Writesonic dropped them from its current API reference, and only unlisted legacy pages last updated in March 2025 still describe them. Starting fix generation is AI work run on your Writesonic plan; the docs state no per-call credit cost. Every other tool only reads results already computed (no tool starts a site audit or a tracking run). Authentication is an API key from the Writesonic API Dashboard (avatar menu, API Dashboard, toggle the API on, Reveal API Key), sent in the X-API-Key header; API access is included only on Writesonic plans that list it.

Connect Writesonic How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Writesonic.Built from the 19 Writesonic tools below, on the connection you add once.

What you can do with Writesonic

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Prompt, 2 operations

list_prompt_performanceList prompt performanceGETPer prompt, topic and platform aggregates for one date: models, mentions, average visibility score and average rank. Pass compare_date to get the *_compare and *_diff fields.
list_promptsList promptsGETLists the prompts (queries) configured for a site, with their topic, country code and status. Prompt ids filter the performance and citation exports. Paginated with page and size.

Recommendation, 2 operations

generate_page_fixesGenerate page fixesPOSTStarts AI fix generation for one page and one issue type (AI work on your plan, no per-call cost documented); returns at once with success and a message. Poll get_fix_generation_status, then read get_page_fixes.
get_page_fixesGet page fixesGETReturns the generated fixes for one page and one issue type: current, suggested and final value, reasoning and state. Run generate_page_fixes first and wait until get_fix_generation_status says COMPLETED.

All, 1 operation

list_page_fixesList page fixesGETReturns every generated fix for one page, grouped by issue type, with each group's generation status and error message.

Answer, 1 operation

list_ai_answersList ai answersGETRaw AI answers for one date, one row per response: prompt, topic, platform, the answer text, rank, visibility score and the brand mentions found in it. Not aggregated.

Bulk, 1 operation

bulk_generate_page_fixesBulk generate page fixesPOSTStarts AI fix generation for every supported issue type on several pages at once (AI work on your plan, no per-call cost documented); returns a success flag and message per page. Then read list_page_fixes.

Citation, 1 operation

list_content_citationsList content citationsGETLists the pages AI answers cited on one date (title, URI, domain), with the websites mentioned and the responses that cited them. Pass compare_date for the *_compare fields.

Get issue, 1 operation

list_seo_audit_issuesList seo audit issuesGETReturns the latest SEO audit issues of a site, grouped by issue class (content, meta, link issues) and category. severity is required: Critical, Warning, or both.

Issue, 1 operation

get_page_issuesGet page issuesGETReturns one page's content refresh analysis: citation counts, cited websites and its issues (issue_id, issue_type, heading, reasoning, priority 1 high to 3 low). Issue ids can be excluded in generate_page_fixes.

Issues summary, 1 operation

get_issues_summaryGet issues summaryGETSummarises content refresh issues across all pages of a site: per issue type, its heading, how many URLs it impacts and which, plus the last scan date.

Keyword, 1 operation

list_content_keywordsList content keywordsGETLists the keywords seen in AI answers on one date, with theme, subtheme and the responses using them. Pass compare_date for the *_compare fields.

Page, 1 operation

list_content_refresh_pagesList content refresh pagesGETLists the site's content refresh pages: title, source_url, type, citation counts, cited websites, issues and optimization status. Pass a page's source_url as page_url to the other content refresh tools.

Platform, 1 operation

list_platformsList platformsGETLists the AI platforms (models) enabled for a site: id, name, provider and category. Platform ids filter the performance and content exports. Paginated with page and size.

Scanned url, 1 operation

list_scanned_urlsList scanned urlsGETLists the URLs of the site that the content refresh scan has analysed, with title and last scan time.

Status, 1 operation

get_fix_generation_statusGet fix generation statusGETReturns the generation status of the fixes for one page and one issue type (status, error_message). Call after generate_page_fixes until the status is final, then read them with get_page_fixes.

Summary, 1 operation

get_performance_summaryGet performance summaryGETDaily AI search KPIs per website for one date: prompts, results, mentions, visibility score and rank, averaged across prompts and platforms. Pass compare_date to get deltas.

Topic, 1 operation

list_topicsList topicsGETLists the topics tracked for a site (id, name, created and updated times). Topic ids filter the performance and content exports. Paginated with page and size.

Website, 1 operation

list_websitesList websitesGETLists every website tracked for a site: your own properties and the direct and indirect competitors (is_self_website, is_competitor). Website ids filter get_performance_summary.

Use Writesonic from Claude Code, Cursor or Codex

Connect Writesonic in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Writesonic operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Writesonic

Create your workspace
Sign up free. Nothing to install.
Connect Writesonic once
Add your Writesonic connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.