Tool tree
INTENT is a personal intent compiler for the agent-native web. Humans express outcomes. Agents handle coordination.
Implementation
gradeNot graded
gradeNot graded
create_intentAction
Compiles a natural language outcome request into a structured Universal Intent Contract.
View tool JSON
{
"name": "create_intent",
"kind": "act",
"impl": "imperative",
"description": "Compiles a natural language outcome request into a structured Universal Intent Contract.",
"inputSchema": {
"type": "object",
"properties": {
"intent": {
"type": "string",
"description": "The natural language outcome requirement (e.g. flight, product purchase)."
}
},
"required": [
"intent"
]
}
}run_intentAction
Executes the multi-agent coordination lifecycle (Discovery, Capability Inspection, Analysis, Multi-Turn Negotiation, Outcome Selection).
View tool JSON
{
"name": "run_intent",
"kind": "act",
"impl": "imperative",
"description": "Executes the multi-agent coordination lifecycle (Discovery, Capability Inspection, Analysis, Multi-Turn Negotiation, Outcome Selection).",
"inputSchema": {
"type": "object",
"properties": {
"intentId": {
"type": "string",
"description": "The unique ID of the compiled intent contract to run."
}
},
"required": [
"intentId"
]
}
}approve_and_executeSensitive Action
Authorizes and executes the selected negotiated outcome, locking inventory and generating the final immutable audit receipt.
View tool JSON
{
"name": "approve_and_execute",
"kind": "transact",
"impl": "imperative",
"description": "Authorizes and executes the selected negotiated outcome, locking inventory and generating the final immutable audit receipt.",
"inputSchema": {
"type": "object",
"properties": {
"intentId": {
"type": "string",
"description": "The intent ID being authorized."
},
"approved": {
"type": "boolean",
"description": "Boolean flag indicating explicit human approval."
}
},
"required": [
"intentId",
"approved"
]
}
}