@sly_ai/mcp-server package exposes the full Sly API as MCP tools. Install it into Claude Desktop, Cursor, or any MCP-compatible client and the agent instantly has access to accounts, agents, wallets, transfers, streams, protocols, and more.
Install
Claude Desktop:Remote HTTP MCP
If your runtime prefers HTTP over stdio:Tool categories
Tools are grouped by resource. All respect the calling agent’s KYA tier and wallet policy.Accounts
| Tool | Purpose |
|---|---|
sly_create_account | Create person or business account |
sly_list_accounts | List accounts (paginated) |
sly_get_account | Get account by ID |
sly_update_account | Update metadata or contact |
Agents
| Tool | Purpose |
|---|---|
sly_create_agent | Create agent under a parent account |
sly_list_agents | List agents |
sly_get_agent | Get agent by ID |
sly_delete_agent | Delete / revoke |
sly_verify_agent | Start verification flow for a tier upgrade |
sly_get_agent_limits | Read effective KYA limits |
sly_get_agent_transactions | Paginated history |
Wallets
| Tool | Purpose |
|---|---|
sly_create_wallet | Create an account wallet |
sly_list_wallets | List wallets |
sly_get_wallet | Get wallet by ID |
sly_get_wallet_balance | Balance available + pending |
sly_wallet_deposit | Internal deposit (from another Sly wallet) |
sly_wallet_withdraw | Withdraw to external rail |
sly_wallet_test_fund | Sandbox faucet |
Agent wallets + policies
| Tool | Purpose |
|---|---|
sly_agent_wallet_get | Get agent wallet |
sly_agent_wallet_fund | Fund from parent account |
sly_agent_wallet_set_policy | Update policy |
sly_agent_wallet_evaluate_policy | Dry-run a spending decision |
sly_agent_wallet_get_evaluations | Audit log of past evaluations |
sly_agent_wallet_get_exposures | Projected future spend |
sly_agent_wallet_freeze | Kill-switch (spending) |
sly_agent_wallet_unfreeze | Resume |
Transfers & settlement
| Tool | Purpose |
|---|---|
sly_create_settlement | Create a settlement |
sly_get_settlement_quote | FX quote |
sly_get_settlement_status | Poll settlement |
Protocol tools
UCP:sly_ucp_create_checkout, sly_ucp_get_checkout, sly_ucp_update_checkout, sly_ucp_complete_checkout, sly_ucp_cancel_checkout, sly_ucp_list_checkouts, sly_ucp_batch_checkout, sly_ucp_batch_complete, sly_ucp_add_payment_instrument, sly_ucp_add_fulfillment_event, sly_ucp_cancel_order, sly_ucp_update_order_status, sly_ucp_get_order, sly_ucp_list_orders, sly_ucp_discover
ACP: sly_acp_create_checkout, sly_acp_get_checkout, sly_acp_complete_checkout, sly_acp_list_checkouts, sly_acp_batch_checkout
AP2: sly_ap2_create_mandate, sly_ap2_get_mandate, sly_ap2_list_mandates, sly_ap2_update_mandate, sly_ap2_execute_mandate, sly_ap2_cancel_mandate
x402: sly_x402_create_endpoint, sly_x402_get_endpoint, sly_x402_list_endpoints, sly_x402_pay, sly_x402_verify, sly_agent_x402_sign
A2A: sly_a2a_send_task, sly_a2a_get_task, sly_a2a_list_tasks, sly_a2a_discover_agent
MPP: sly_mpp_open_session, sly_mpp_get_session, sly_mpp_list_sessions, sly_mpp_close_session, sly_mpp_pay, sly_mpp_verify_receipt, sly_mpp_list_transfers
Governance + compliance
| Tool | Purpose |
|---|---|
sly_explain_rejection | Explain why a prior attempt was blocked |
sly_escalate_to_human | Route to approval workflow |
sly_request_limit_increase | Ask for a temporary limit bump |
sly_open_dispute | Open a transaction dispute |
Agent funding (sandbox helpers)
| Tool | Purpose |
|---|---|
sly_agent_fund_eoa | Fund agent’s externally-owned account |
sly_agent_refill_faucet | Refill sandbox faucet |
sly_agent_evm_key_provision | Provision a fresh EVM keypair for the agent |
Example invocation
Permission filtering
The MCP server only exposes tools the calling agent/credentials can actually invoke. An agent at KYA Tier 0 won’t see tools that require Tier 2 even if they’re installed — reducing confusion for the LLM and preventing wasted tool calls.Scanner MCP (separate server)
The Scanner service has its own MCP server with 19 tools for merchant readiness intelligence:scan_merchant, batch_scan, get_readiness_report, get_protocol_adoption, get_heat_map, find_best_prospects, compare_merchants, generate_snapshot, and run_agent_shopping_test.
For agent authors
If you’re building an agent that uses MCP tools: the server automatically translates the LLM’s natural-language tool calls into Sly API requests. You don’t write HTTP code — you write good prompts, and the model picks the right tool from the catalog. Good practice:- Give the agent your intended purpose in the system prompt (e.g. “You help pay supplier invoices. Use the Sly tools to check balances and send transfers.”)
- The agent will autonomously pick among the tools
- Wallet policies enforce what’s actually allowed — so even a confused LLM can’t send a 500/day cap
