Teller MCP – Borrow USDC & Altcoins (no margin calls)

Expose the Teller delta-neutral + lending Model Context Protocol server. Use this when you need to install, run, or update the Tellermcp MCP backend so agent...

This skill appears to be what it says: an MCP server that queries Teller's public Delta-Neutral API and returns data or encoded on-chain transactions. Before installing, review and agree to the following: 1) The server will make outbound requests to the configured Teller API (default https://delta-neutral-api.teller.org). If you don't trust that endpoint, set TELLER_API_BASE_URL to a safe host or run a local mock. 2) The tools produce encoded transaction calldata — the skill does not hold or require private keys, but executing returned transactions with your wallet/private key can move funds. Never supply private keys or secrets to the skill. 3) Run npm install/build in a sandbox or CI runner if you want to audit dependencies (package-lock.json is included). 4) If you plan to expose the server in production, validate dependencies (especially @modelcontextprotocol/sdk) and consider network and process isolation. Overall the package is coherent with its stated purpose; proceed if you trust the upstream API and are comfortable running a Node MCP server that can produce on-chain transaction builders.

Type: OpenClaw Skill Name: teller-borrow Version: 0.2.2 The OpenClaw skill 'teller-borrow' is a Model Context Protocol (MCP) server that exposes Teller delta-neutral and lending API functionalities. It makes outbound HTTP/HTTPS requests to `https://delta-neutral-api.teller.org` (configurable via `TELLER_API_BASE_URL` environment variable), which is its stated purpose. The skill utilizes `zod` for robust input validation on all API parameters, including strict regex checks for Ethereum addresses, mitigating potential injection risks. Communication with the OpenClaw agent occurs securely via `StdioServerTransport`. There is no evidence of intentional harmful behavior such as data exfiltration, unauthorized command execution, persistence mechanisms, or malicious prompt injection attempts in SKILL.md. All observed behaviors are consistent with a legitimate API wrapper.