delete-agent

Delete an agent from the project's agent registry by its agent_id. Needs the fabric:orchestrate.write grant.

curl -sS -X POST https://fabric.wexa.ai/mcp/$FABRIC_PROJECT_ID \
  -H "authorization: Bearer $FABRIC_API_KEY" \
  -H "Mcp-Session-Id: $MCP_SESSION_ID" \
  -H "content-type: application/json" -H "accept: application/json" \
  -d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"delete-agent","arguments":{"agent_id":"6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48"}}}'

{
  "content": [
    {
      "type": "text",
      "text": "{\"lifecycle_id\":\"qlc_000208\",\"result\":{\"agent_id\":\"6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48\",\"container_remains\":false,\"deleted\":true}}"
    }
  ],
  "isError": false
}

curl -sS https://fabric.wexa.ai/v1/delete-agent \
  -H "authorization: Bearer $FABRIC_API_KEY" \
  -H "content-type: application/json" \
  -d '{"agent_id":"6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48"}'

{
  "lifecycle_id": "qlc_000208",
  "result": {
    "agent_id": "6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48",
    "container_remains": false,
    "deleted": true
  }
}

import { Fabric } from '@wexa-fabric/sdk'

const fabric = new Fabric()
const result = await fabric.deleteAgent({
  agent_id: '6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48'
})

// result is the payload itself, with no `.result` to read
{
  agent_id: '6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48',
  container_remains: false,
  deleted: true
}

from wexa import Fabric

fabric = Fabric()
out = fabric.delete_agent(
    agent_id="6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48",
)

# out is the result, already unwrapped
{
    "agent_id": "6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48",
    "container_remains": False,
    "deleted": True
}

Parameters

Parameter	Type	Required	Notes
`agent_id`	string	Required	Agent id from `list-agents` or `create-agent`. An agent in this project.

Show full response

A real response, captured from a live call to delete-agent:

{
  "content": [
    {
      "type": "text",
      "text": "{\"lifecycle_id\":\"qlc_000208\",\"result\":{\"agent_id\":\"6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48\",\"container_remains\":false,\"deleted\":true}}"
    }
  ],
  "isError": false
}

{
  "lifecycle_id": "qlc_000208",
  "result": {
    "agent_id": "6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48",
    "container_remains": false,
    "deleted": true
  }
}

// the object you get back IS the result — there is no `.result` to read
{
  agent_id: '6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48',
  container_remains: false,
  deleted: true
}

// The TS SDK names these camelCase — lifecycleId, traceId — where Python uses
// lifecycle_id and trace_id. Both are non-enumerable: readable, but omitted by
// Object.keys() and JSON.stringify(). `result.lifecycleId` here is undefined.
result.lifecycleId // 'qlc_000208'

# a dict subclass; `result` is already unwrapped
{
 "agent_id": "6ab196817566f41b5c28e335.6ab19681c73ec18cad7c9c48",
 "container_remains": False,
 "deleted": True
}

out.lifecycle_id  # 'qlc_000208' — an attribute, not a key

More details

Afterwards get-agent on that id reports not found. The container the agent was stored in goes with it, so nothing is left behind and no second call is needed. If that part fails the result carries container_remains: true, and delete-process-flow removes it. For a multi-agent process flow use the process-flow tools instead.

Every tool call works the same way: How every tool call works.