This is an intent execution engine that turns high-level agent commands into Base L2 transactions. It exposes four MCP tools: submit intents for transfers, swaps, or multi-hop settlements, check execution status by intent ID, pull aggregate stats on success rates and volume, and list available execution providers with reliability scores. Behind the scenes it routes to providers, logs executions to SQLite, and handles atomic settlement. Part of the Hive Civilization agent economy stack. Reach for this when you're building autonomous agents that need to move USDC or execute DeFi operations without manually constructing transactions.
HiveExecute is a Model Context Protocol (MCP) server that resolves and executes agent intents on Base L2. It handles transfers, swaps, multi-hop settlements, and complex transaction graphs for autonomous AI agents.
POST /mcp
All tool interactions use JSON-RPC 2.0 over HTTP.
{
"jsonrpc": "2.0",
"id": 1,
"method": "initialize"
}
{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/list"
}
{
"jsonrpc": "2.0",
"id": 3,
"method": "tools/call",
"params": {
"name": "hiveexecute_submit_intent",
"arguments": {
"intent_type": "transfer",
"from_did": "did:hive:alice",
"to_did": "did:hive:bob",
"amount_usdc": 10.00,
"memo": "Payment for services"
}
}
}
| Tool | Description |
|---|---|
hiveexecute_submit_intent | Submit an intent for resolution and execution |
hiveexecute_get_status | Get execution status by intent ID |
hiveexecute_get_stats | Get execution statistics (total executions, success rate, volume) |
hiveexecute_list_providers | List available execution providers and capabilities |
hiveexecute_submit_intentSubmit an intent for atomic resolution and execution.
Parameters:
| Param | Type | Required | Description |
|---|---|---|---|
intent_type | string | Yes | transfer, swap, or multi_hop |
from_did | string | Yes | Sender DID |
to_did | string | Yes | Recipient DID |
amount_usdc | number | Yes | Amount in USDC |
memo | string | No | Description or memo |
hiveexecute_get_statusGet the execution status of a previously submitted intent.
Parameters:
| Param | Type | Required | Description |
|---|---|---|---|
intent_id | string | Yes | Execution ID from submit_intent |
hiveexecute_get_statsReturns total executions, success rate, total volume in USDC, total savings, and daily execution count. No parameters.
hiveexecute_list_providersReturns available execution providers grouped by intent type, with reliability scores and execution history. No parameters.
In addition to the MCP endpoint, HiveExecute exposes a REST API:
| Method | Path | Description |
|---|---|---|
POST | /v1/execute_intent | Execute an intent atomically |
GET | /v1/execute_intent/history/{did} | Execution history for an agent |
GET | /v1/execute_intent/stats | Execution statistics |
GET | /v1/execute_intent/providers | Available providers by intent type |
GET | /health | Health check |
Node.js with Express. SQLite for execution logs and provider scores. Part of the Hive Civilization autonomous agent economy on Base L2.
Proprietary — Hive Civilization
Hive Civilization is the cryptographic backbone of autonomous agent commerce — the layer that makes every agent transaction provable, every payment settable, and every decision defensible.
This repository is part of the PROVABLE · SETTABLE · DEFENSIBLE pillar.
io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent