Skip to main content
GET
/
.well-known
/
agent-card.json
Agent card discovery
curl --request GET \
  --url https://mesh.svantic.com/.well-known/agent-card.json
{
  "name": "savant",
  "description": "Savant AI Agent Framework — orchestrates intelligent task execution with planning, learning, and tool integration.",
  "url": "https://mesh.example.com/send",
  "version": "2.0.0",
  "protocolVersion": "0.2.3",
  "defaultInputModes": [
    "application/json",
    "text/plain"
  ],
  "defaultOutputModes": [
    "application/json",
    "text/plain"
  ],
  "capabilities": {
    "streaming": true,
    "pushNotifications": false,
    "stateTransitionHistory": false
  },
  "skills": [
    {
      "id": "orchestration",
      "name": "Task Orchestration",
      "description": "Routes user requests to specialized agents: chat, executor, planner, code repair, document processing.",
      "tags": [
        "core",
        "orchestration"
      ]
    },
    {
      "id": "planning",
      "name": "Task Planning",
      "description": "Breaks complex tasks into execution plans with dependency ordering and agent mesh construction.",
      "tags": [
        "core",
        "planning"
      ]
    },
    {
      "id": "transcript-link:process_document",
      "name": "transcript-link — Document Processing",
      "description": "Processes academic transcripts and employment documents.",
      "tags": [
        "document",
        "remote:transcript-link",
        "instances:3"
      ],
      "examples": [
        "Process the transcript at https://example.com/doc.pdf",
        "Extract GPA from this academic record"
      ],
      "parameters": {
        "type": "object",
        "properties": {
          "document_url": {
            "type": "string",
            "description": "URL of the document to process."
          },
          "document_type": {
            "type": "string",
            "description": "Type of document.",
            "enum": [
              "transcript",
              "pay_stub",
              "tax_return"
            ]
          },
          "output_format": {
            "type": "string",
            "description": "Desired output format.",
            "enum": [
              "json",
              "pdf"
            ]
          }
        },
        "required": [
          "document_url",
          "document_type"
        ]
      }
    }
  ],
  "provider": {
    "organization": "Svantic",
    "url": "https://svantic.ai"
  },
  "extensions": {
    "instances": {
      "transcript-link": 3
    }
  }
}

Documentation Index

Fetch the complete documentation index at: https://docs.svantic.com/llms.txt

Use this file to discover all available pages before exploring further.

Response

A2A agent card document.

A2A-protocol-compliant agent card. Describes the agent's identity, capabilities, skills, supported transports, and security requirements. Conforms to the AgentCard definition in the A2A specification.

​
name
string
required

A human-readable name for the agent.

​
description
string
required

A human-readable description of the agent's purpose.

​
url
string<uri>
required

The preferred endpoint URL for interacting with the agent (JSON-RPC).

​
version
string
required

The agent's own version number.

​
protocolVersion
string
required

The version of the A2A protocol this agent supports (e.g. "0.2.3").

​
defaultInputModes
string[]
required

Default set of supported input MIME types for all skills.

​
defaultOutputModes
string[]
required

Default set of supported output MIME types for all skills.

​
capabilities
object
required

Capabilities supported by the agent.

​
skills
object[]
required

The set of skills (capabilities) the agent can perform.

​
provider
object

Information about the agent's service provider.

​
preferredTransport
string

Transport protocol for the preferred endpoint. Defaults to "JSONRPC" when not specified.

​
additionalInterfaces
object[]

Additional transport/URL combinations the agent supports, enabling transport negotiation and fallback.

​
security
object[]

Security requirement objects (OpenAPI 3.0 style). Each object lists security schemes that can be used. The list represents a logical OR of ANDs.

​
securitySchemes
object

Security scheme definitions available for authorization. Follows the OpenAPI 3.0 Security Scheme Object.

​
signatures
object[]

JSON Web Signatures computed for this AgentCard.

​
supportsAuthenticatedExtendedCard
boolean
default:false

If true, the agent provides an extended agent card with additional details to authenticated users.

​
iconUrl
string<uri>

An optional URL to an icon for the agent.

​
documentationUrl
string<uri>

An optional URL to the agent's documentation.

​
extensions
object

Svantic-specific extensions. Currently includes instances — a map of agent_type to instance count.