Elfa AI
About
Elfa is our flagship finance agent, powered by our real-time intelligence stack (Iris). Iris synthesises signals across social, on-chain, and off-chain market data, then traces their second and third-order effects, so agents act on greater context.
Details
- Author
- elfa-ai
- Categories
- Finance, Other
Jump to
Setup
Install Elfa AI in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/elfa-ai/mcp
Follow the installation instructions in the repository README, then restart your MCP client.
Model Context Protocol server for theElfa API— crypto social intelligence from X and Telegram, plusAuto, a condition engine that watches the market and fires an action when your conditions are met.
Works with any MCP client: Claude Code, Claude Desktop, Cursor, VS Code, Codex, and anything else that speaks MCP.
Get an API key atdev.elfa.ai. No install step —npxfetches the server on demand.
Downloadelfa-mcp-<version>.mcpbfrom thelatest releaseand open it. Claude Desktop installs it, prompts for your API key, and keeps it updated. Nothing else to configure.
claude mcp add elfa --env ELFA_API_KEY=your-key -- npx -y @elfa-ai/mcp
Cursor, VS Code, Claude Desktop, and other clients
{ "mcpServers": { "elfa": { "command": "npx", "args": ["-y", "@elfa-ai/mcp"], "env": { "ELFA_API_KEY": "your-key" } } } }
VS Code uses"servers"instead of"mcpServers". Everything else is the same.
Ask"what's trending in crypto right now?"to confirm it works.
The timeout is high and retries are off on purpose. The interpretation endpoints are LLM-backed and can take over a minute, and they cost credits per attempt, so a silent retry would bill you again for a call you never saw. RaiseELFA_RETRIESonly if you are calling the cheap measurement endpoints.
Some MCP clients apply their own timeout, often around 60 seconds.narrativesandmarket_chatcan exceed that; the request still completes and is still charged, even if the client gives up first.
WithoutELFA_HMAC_SECRETnotification-only Auto mutations still work. Anything else returns a message telling you what to set.
11 tools, mapped to every documented/v2operation.
- chat-stream-v2— A tool call returns one result, so streaming adds nothing. market_chat covers the same analysis.
- auto-stream-queries-v2— Long lived streams have no tool equivalent. Poll with auto_query.
- auto-stream-query-v2— Long lived streams have no tool equivalent. Poll with auto_query.
Streaming endpoints stay available through theSDKsfor applications that can consume SSE.
The tools deliberately do not inherit every API default, because an agent pays for verbosity in context.
Every value is still settable per call, andpageSizeaccepts up to 100.
Auto queries run unattended. Once armed, a query keeps evaluating and fires its action without asking again.
- auto_build— describe what to watch in plain language, get EQL back
- auto_validate— check the syntax and get the credit cost
- auto_query_write— activate it
Actions can notify you, call a webhook, message a Telegram bot, or run an LLM analysis.
There is no push channel over MCP. Pollauto_querywithmethod=get, and wait for the returnedpollAfterSecondsbetween calls.
The same server runs over Streamable HTTP for hosted deployments:
ELFA_MCP_TRANSPORT=http ELFA_MCP_PORT=3000 npx -y @elfa-ai/mcp
It is stateless — no sessions, one server instance per request, safe behind a load balancer. Credentials come fromx-elfa-api-keyandx-elfa-hmac-secretrequest headers, falling back to the environment. SetELFA_MCP_ALLOWED_ORIGINSto a comma separated allowlist when exposing it publicly.
api_statusis the fastest way to tell an auth problem from a credit problem.
Mentions, news and narratives return third-party social text that anyone can write. The server marks it as untrusted in every response, and the server instructions tell the model to treat it as data. Keep that in mind before letting an agent chain from that content intoauto_query_write.
npm install npm run build npm run verify
npm run verifyruns typecheck, tests, the spec drift check, and the docs check.
manifest.jsonmaps every documented API operation to the tool that covers it.npm run check:driftfails if the API grows an operation the server does not handle. The tool table above is generated from the same file withnpm run docs:tools.
Data and execution layer for tokenized real-world assets (RWAs) on Injective. Open catalog tools expose indexed assets, live market data, price history, routable venues, and a 6-dimension risk score (backing, enforceability, control, exit, liquidity, social) for every asset. Trading tools (buy, sell, deposit, withdraw) work via lazy OAuth — the first trade call triggers sign-in with Realmint and a one-time wallet delegation; trades are then signed server-side under a spending cap, with no API keys or seed phrases in the client. Also supports x402 pay-per-call funding for agents.
Schedule and queue text, image, and video posts to many social networks, including LinkedIn, X, Bluesky, Instagram, Facebook, Threads, Tiktok.
An MCP server to connect and interact with all your social media
Streamline your social media with Postiz. Schedule posts, analyze performance, and manage all accounts in one place
Installable MCP server for AgentLux marketplace, identity, creator, services, social, and Base/x402 commerce flows.
Draft, schedule and publish social posts to LinkedIn, X, Instagram, Facebook, Threads, TikTok, Pinterest and YouTube from your AI assistant, in a brand voice learned per account.
AI-powered social media posting across 14 platforms with per-platform content adaptation, scheduling, and 30-day calendar generation
Schedule, generate and publish social posts to X, LinkedIn, Instagram, Threads and YouTube from any MCP client.
Create, schedule, and publish social media posts to 9 platforms from Claude, Cursor, or any AI agent. Hosted server with OAuth or API key auth.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
