An artificial intelligence agent — a software system designed to autonomously or semi-autonomously take actions, answer questions, and complete tasks on behalf of a human or organisation, operating within a defined capability boundary and accountability chain.
ajp:agentIdTextrequiredUnique identifier for this agent instance
ajp:agentVersionTextrequiredVersion string (semantic versioning recommended)
ajp:agentPurposeTextrequiredPlain-English primary purpose statement
ajp:capabilityDeclarationajp:CapabilityDeclarationrequiredStructured capability and prohibition record
ajp:trainingDataSources[ajp:TrainingDataSource]requiredProvenance of the agent's knowledge base
ajp:modelConfigurationajp:ModelConfigurationrequiredTechnical configuration of the AI model
ajp:agentPassportajp:AgentPassportrequiredCryptographic identity credential
ajp:systemPromptHashTextrequiredSHA-256 fingerprint of the agent system prompt
ajp:agentBuilderschema:OrganizationrequiredThe verified organisation that built this agent
ajp:agentOwnerschema:OrganizationrequiredThe organisation that owns the agent IP
ajp:deploymentConstraintsajp:DeploymentConstraintsSafety and operational deployment limits
ajp:humanApprovalGates[ajp:HumanApprovalGate]Actions requiring human sign-off before execution
ajp:maxTransactionValueschema:MonetaryAmountMaximum financial value the agent can authorise
ajp:agentPersonaajp:AgentPersonaThe agent's presented identity to users
ajp:auditLogajp:AgentAuditRecordLink to public audit record
ajp:knowledgeBoundariesajp:KnowledgeBoundaryScope and limits of agent knowledge
{
"@context": {
"@vocab": "https://schema.org/",
"ajp": "https://schema.anthonyjamespeacock.com/"
},
"@type": [
"ajp:AIAgent"
],
"ajp:agentId": "<Text>",
"ajp:agentVersion": "<Text>",
"ajp:agentPurpose": "<Text>"
}