> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mindset.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Agents API

> Parameters full description

<Card title="Note" icon="link" iconType="solid" href="/deploy/api/agent-management/retrieve-agents" cta="Agents API playground">
  You can find below the full description of Parameters for each POST/PATCH/GET/DELETE methods and also what parameters you can expect from the API responses.

  <br />

  If you want to test the API, please see the Agents API playground.
</Card>

<AccordionGroup>
  <Accordion title="POST - Create a new agent">
    ## Parameters:

    #### **agentName** *String*

    Name of the agent

    #### **agentPurpose** *String*

    Purpose of the agent

    Please read [Agent creation - Provide clear purpose](/sdk-api/sdk/sdk2-best-practices#1-provide-clear-purpose)

    #### **formatting** *String*

    Output formatting instructions

    Please read [Agent creation - Specify clear formatting rules](/sdk-api/sdk/sdk2-best-practices#3-specify-clear-formatting-rules)

    #### **agentPersonality** *String*

    Personality of the agent

    Please read [Agent creation - Define personality traits](/sdk-api/sdk/sdk2-best-practices#2-define-personality-traits)

    #### **agentPolicyRules** *String*

    Policy rules for the agent

    Please read [Agent creation - Set appropriate policies](/sdk-api/sdk/sdk2-best-practices#4-set-appropriate-policies)

    #### **contextUids** *String\[]*

    Array of existing context Uids (max 30)

    #### **mcpserverUids** *String\[]*

    Array of existing mcpserver Uids (max 5)

    #### **planUids** *String\[]*

    Array of existing plan UIDs (max 20)

    #### **skillUids** *String\[]*

    Array of existing skill UIDs (max 20)

    #### **tags** *String\[]*

    Array of tags (max 10). You will be able to view reports based grouping or filtering of activity according to the tags applied when creating the agent.

    #### **access** *String*

    default : `restrictedAccess` Access control mode for this agent. Use Agent Sessions with restrictedAccess agents.

    Possible values: \[`restrictedAccess`, `openAccess`]

    #### **externalTenantId** *String*

    External tenant identifier. Valid characters are A-Za-z0-9\_-.

    #### **iceBreakerMessage** *String*

    Invisible instruction to prompt the agent to greet users in a specific way or style (max 1000)
  </Accordion>

  <Accordion title="PATCH - Update an existing agent">
    ## Parameters:

    #### **agentName** *String*

    Name of the agent

    #### **agentPurpose** *String*

    Purpose of the agent

    #### **formatting** *String*

    Output formatting instructions

    #### **agentPersonality** *String*

    Personality of the agent

    #### **agentPolicyRules** *String*

    Policy rules for the agent

    #### **live** *Boolean*

    Set the agent status: live or not live

    #### **contextUids** *String\[]*

    Array of existing context Uids (max 30)

    #### **mcpserverUids** *String\[]*

    Array of existing mcpserver Uids (max 5)

    #### **planUids** *String\[]*

    Array of existing plan UIDs (max 20)

    #### **skillUids** *String\[]*

    Array of existing skill UIDs (max 20)

    #### **tags** *String\[]*

    Array of tags (max 10). You will be able to view reports based grouping or filtering of activity according to the tags applied when creating the agent.

    #### **access** *String*

    default : `restrictedAccess` Access control mode for this agent. Use Agent Sessions with restrictedAccess agents.

    Possible values: \[`restrictedAccess`, `openAccess`]

    #### **externalTenantId** *String*

    External tenant identifier. Valid characters are A-Za-z0-9\_-.

    #### **iceBreakerMessage** *String*

    Invisible instruction to prompt the agent to greet users in a specific way or style (max 1000)
  </Accordion>

  <Accordion title="GET - Query the list of agents">
    ## Parameters:

    #### **startAfter - Querying for “the next page”** *String*

    Use the `startAfter` parameter and provide the `createdAt` string of the last agent in the current results set When there are no more results you will receive an empty array in the response and a 404 status code

    #### **contextUid** *String*

    Filter results to a specific contextUid (cannot be used with mcpserverUid)

    #### **mcpserverUid** *String*

    Filter results to a specific mcpserverUid (cannot be used with contextUid)

    #### **externalTenantId** *String*

    Filter results to a specific externalTenantId
  </Accordion>

  <Accordion title="DELETE - Delete an agent by providing its uid" />

  <Accordion title="RESPONSE DATA">
    You will get the same response data when calling POST, PATCH or GET.

    ***Note***: When retrieving a collection of agents, the response will be an array of objects.

    #### **uid** *String*

    Mindset internal agent uid

    #### **agentName** *String*

    Name of the agent

    #### **agentPurpose** *String*

    Purpose of the agent

    #### **formatting** *String*

    Output formatting instructions

    #### **agentPersonality** *String*

    Personality of the agent

    #### **agentPolicyRules** *String*

    Policy rules for the agent

    #### **live** *Boolean*

    Agent status: live or not live

    #### **createdAt** *ISO timestamp*

    Agent creation date

    #### **contextUids** *String\[]*

    Array of existing context Uids (max 30)

    #### **mcpserverUids** *String\[]*

    Array of existing mcpserver Uids (max 5)

    #### **planUids** *String\[]*

    Array of plan UIDs (max 20)

    #### **skillUids** *String\[]*

    Array of skill UIDs (max 20)

    #### **tags** *String\[]*

    Array of tags (max 10).

    #### **access** *String*

    Access control mode for this agent. Possible values: \[`restrictedAccess`, `openAccess`]

    #### **externalTenantId** *String*

    #### **iceBreakerMessage** *String*
  </Accordion>
</AccordionGroup>