Beamswap MCP
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BEAMSWAP_API_URL | No | Defaults to https://api.beamswap.io. Change only to an API you trust. | https://api.beamswap.io |
| BEAMSWAP_WALLET_KEY | No | Optional for connection/free lookups; required for local paid calls and wallet sign-in. Use a separate wallet with a small USDC balance on Base. Never paste a private key into a chat, prompt, issue or shared config. | |
| BEAMSWAP_SESSION_TOKEN | No | Optional existing wallet session. session_create can create one and retain it in process memory. | |
| BEAMSWAP_PAYMENT_TREASURY | No | Advanced self-hosting only. Defaults to Beamswap's treasury. Changing the API URL does not change this payment recipient. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| session_createA | Create a 24-hour session for the configured wallet. Free. The session unlocks GLINT tier quota, discounts and execution fees and remains only in this MCP process. |
| token_infoA | Base ERC-20 facts: metadata, supply, deployer, USD price, round-trip liquidity check. Costs $0.005 USDC via x402. |
| portfolioB | ETH + ERC-20 balances of a Base address with USD values. Costs $0.02 USDC via x402. |
| swap_quoteB | Best swap output across Base aggregators (quote only, no calldata). Costs $0.005 USDC via x402. |
| swap_routeA | Best swap route across Base aggregators as ready-to-sign calldata. Costs $0.01 USDC via x402 plus a 10 bps fee inside the route (less for GLINT tiers). Returns to/data/value; you sign and send it. |
| watch_createB | Watch Base balances or USD prices; signed webhook on trip. $0.01 per item per day via x402. |
| watch_getA | Read one watch: its condition, status, last observed value and recent webhook deliveries. Free. |
| watch_deleteA | Stop one watch. Free, idempotent, and never refunded: the days were bought up front. |
| distribution_createA | Deploy a cumulative Merkle token distributor on Base and return approval, gross funding and unpause transactions for the owner to send. Costs $50 USDC via x402; the API deploys the clone and settles payment, while owner transactions remain unsent. |
| distribution_getA | Read a token distribution, including its root, contract, promised total, funded balance and claim URL. Free. |
| distribution_proofA | Get one recipient cumulative amount and Merkle proof for a distribution. Costs $0.001 USDC via x402; a missing recipient is a free 404. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Each tool targets a distinct resource or action, and the watch and distribution families are clearly separated. The only close pair is swap_quote vs swap_route, but their descriptions distinguish quote-only from ready-to-sign calldata well enough.
Most names follow a resource_first snake_case pattern such as watch_create, distribution_get, and session_create. A few names like portfolio, token_info, and swap_quote use noun-style suffixes instead of verb_action, so the pattern is mostly consistent but not uniform.
With 11 tools, the server is well within the ideal scope and each tool covers a distinct capability: portfolio, token info, swap quoting/routing, watch lifecycle, distribution lifecycle, and session creation. No tool feels redundant.
Core workflows are covered: swap quote/route, watch create/get/delete, and distribution create/get/proof. Minor gaps such as listing watches or distributions and a session status endpoint could improve completeness, but agents can complete the main intended operations.