Integration

Sofya integration

Sofya is a web toolkit for AI agents: search the web and get extracted page content (optionally with an AI answer), fetch up to 10 URLs as clean markdown (web pages, PDF, Office documents), ask an AI question about one page, and run deep multi-source research that returns a cited report. Every call spends Sofya credits (1 credit = $0.0025; only 2xx answers are charged), and the account's balance, top-ups and daily usage can be read back. Not exposed: the MCP transport, agent sign-up, payment-link creation, API key and connected-app management, and account preferences. Authentication is a Sofya API key (dashboard, API keys) sent as a Bearer token.

Connect Sofya How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Sofya.Built from the 10 Sofya tools below, on the connection you add once.

What you can do with Sofya

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.

Balance, 1 operation

get_credit_balanceGet credit balanceGETReturns the account's spendable credits (plan and purchased), when free-tier credits reset, the calling key's monthly cap and use this month, and the top-up rates and dashboard link. Free, no credits.

Extract, 1 operation

extract_from_pageExtract from pagePOSTReads one web page and answers a question about it with AI: content is a free-text answer (not structured JSON). Spends 10 credits, or 1 when the page has too little text (usage.low_content true, content empty).

Fetch, 1 operation

fetch_urlsFetch urlsPOSTFetches 1 to 10 URLs (web pages, PDF, Word, Excel, PowerPoint, EPUB, CSV, XML, text) and returns each as clean markdown in results[]. Spends 1 credit per URL that succeeds; a failed URL is free.

Me, 1 operation

get_accountGet accountGETReturns the account behind the key: credit balance (plan and purchased), free-tier status and reset date, total requests, the owner's email and GitHub username (personal data) and alert settings. Free.

Research, 1 operation

run_researchRun researchPOSTResearches a question across many sources and returns report, a markdown report with citations, plus sources[] and sub_queries[]. Spends 50 credits; usually takes 15 to 25 seconds (55 second limit).

Search, 1 operation

search_webSearch webPOSTSearches the web and returns results[] with title, url, page content and date, plus an optional AI answer. Spends credits: snippets depth 1, basic depth 2, include_answer adds 2. With an answer it can take up to 60 seconds.

Status, 1 operation

get_service_statusGet service statusGETReturns the live health of Sofya's search, fetch, extract, research and MCP services: overall status, each service's status, typical latency and last check, and recent incidents. Free, no credits.

Transaction, 1 operation

list_transactionsList transactionsGETLists the account's last 50 top-ups and adjustments, newest first: type (credit or debit), amount in credits, balance_after, amount_usd_cents and receipt flag. Per-call charges are in get_usage.

Update, 1 operation

list_product_updatesList product updatesGETReturns Sofya's public changelog: updates[], each with date, title, slug and body (markdown) plus body_html. Free, no credits.

Usage, 1 operation

get_usageGet usageGETReturns daily usage rows, newest day first, one per day, endpoint and API key: request_count and total_credits. Covers days 1 to 90 back (default 7); page further back with offset_days. Free.

Use Sofya from Claude Code, Cursor or Codex

Connect Sofya in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Sofya 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 Sofya

Create your workspace
Sign up free. Nothing to install.
Connect Sofya once
Add your Sofya 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.