sapiens-mcp
Enables users to generate images, articles, voice, music, and more from their Sapiens Sintéticos account using natural language, spending credits and saving to their profile.
README
sapiens-mcp
On the official MCP registry as com.sapiensinteticos/sapiens.
An MCP server to operate your Sapiens Sintéticos account from Claude Code, or any MCP client, in your own account. You ask in plain language ("generate an image of this", "write an essay on that") and it does the work, spending your Sinapses (the house credit) and saving to your profile.
Sapiens Sintéticos is an AI prototyping lab. This server is the exoskeleton: image, video, article, voice, music, a personal Repertório (the creative memory the rest reads from), and the community, all from a single conversation.
Account-gated by design
You connect with a Sapiens account (Google or email), no API key, no card. Accounts and login live on the site, never here. No account yet? Create one at sapiensinteticos.com. Every action shows its cost before it runs, and a failed generation is refunded. The remote endpoint is fail-closed: no valid session, nothing runs.
Connect (two ways)
Remote (any MCP client, streamable HTTP). Point your client at:
https://sapiensinteticos.com/api/mcp/mcp
with the header Authorization: Bearer <sessionToken>. Generate the token at sapiensinteticos.com/conectar-claude. Identity is always the bearer, so there is no local login on the remote transport.
Local (Claude Code, npm / stdio).
claude mcp add sapiens -- npx -y sapiens-mcp
Node 18+. The backend URL is built in, nothing to configure. Then log in: open /conectar-claude while signed in, generate the code (XXXX-XXXX, valid for 5 minutes), and run sapiens_meta with action: "login" and code: "XXXX-XXXX". The 30-day token is saved to ~/.sapiens-mcp/session.json.
Works in Claude Code (the tested path), Gemini CLI, Cursor, Antigravity, and any MCP-speaking client. Only the way you add it changes.
What you can ask (and the cost in Sinapses)
| What | Cost |
|---|---|
| Generate an image (to your gallery) | ~400-500 |
| Write an article in the Sapiens voice (to your profile) | 400 |
| Voice / narration (Helen TTS) | 500 |
| Song lyrics / render (Musicator) | 300 / 3000 |
| Persona art | 450 |
| Repertório, list/edit your articles, check balance | free |
The Claude side always warns the cost before spending. Publishing to the editorial blog and the Coluna Sapiens stays with the platform owner, never your account.
Troubleshooting
- "sessionToken expired" / "Sapiens account not connected". The 30-day token lapsed or was never saved. Open /conectar-claude signed in, generate a fresh code, and run
sapiens_meta action=login code=XXXX-XXXX. - A tool or a new capability went missing after an update. The client runs via
npx -y sapiens-mcp(unpinned) and may be stuck on an old cache. Check what is running withsapiens_meta action=version(binary version, latest on npm, andupToDate). IfupToDate:false, clear the npx cache and restart the client. - "Invalid arguments". The message already names the field that is missing or wrong, redo the call with what it asks. Do not repeat the same failing call (3 failures in a row make the client mark the server unreachable for about a minute, an anti-loop breaker).
- Low balance before generating.
sapiens_meta action=credits(oraction=subscriptionfor the per-bucket detail) shows what is left before you spend on image, music, or video.
Privacy
The server only talks to the public Sapiens backend (Convex). Your identity always comes from your login token, never from loose parameters. Each account only touches what is its own.
The house
Sister properties from the same Borderless house:
- sapiensinteticos.com: the studio, and this server's home.
- aitag.app: a curated directory of AI tools (BR and EU). The Sapiens Repertório pulls from this catalog.
- helenai.wtf: Helen, the house's autonomous humanized AI brand (music, chat, comics). The style anchor of Sapiens.
Feito no espírito borderless. Licença MIT. Com ❤️ /conectar-claude.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.