Struct TurnStartParams 

pub struct TurnStartParams {

    pub additional_context: Option<HashMap<String, AdditionalContextEntry>>,
    pub approval_policy: Option<AskForApproval>,
    pub approvals_reviewer: Option<ApprovalsReviewer>,
    pub client_user_message_id: Option<String>,
    pub collaboration_mode: Option<CollaborationMode>,
    pub cwd: Option<String>,
    pub effort: Option<ReasoningEffort>,
    pub environments: Option<Vec<TurnEnvironmentParams>>,
    pub input: Vec<UserInput>,
    pub model: Option<String>,
    pub multi_agent_mode: Option<MultiAgentMode>,
    pub output_schema: Option<Value>,
    pub permissions: Option<String>,
    pub personality: Option<Personality>,
    pub responsesapi_client_metadata: Option<HashMap<String, String>>,
    pub runtime_workspace_roots: Option<Vec<AbsolutePathBuf>>,
    pub sandbox_policy: Option<SandboxPolicy>,
    pub service_tier: Option<String>,
    pub summary: Option<ReasoningSummary>,
    pub thread_id: String,
}

TurnStartParams

JSON schema

{
 "title": "TurnStartParams",
 "type": "object",
 "required": [
   "input",
   "threadId"
 ],
 "properties": {
   "additionalContext": {
     "description": "Optional client-provided context fragments keyed by an opaque source identifier.",
     "type": [
       "object",
       "null"
     ],
     "additionalProperties": {
       "$ref": "#/definitions/AdditionalContextEntry"
     }
   },
   "approvalPolicy": {
     "description": "Override the approval policy for this turn and subsequent turns.",
     "anyOf": [
       {
         "$ref": "#/definitions/AskForApproval"
       },
       {
         "type": "null"
       }
     ]
   },
   "approvalsReviewer": {
     "description": "Override where approval requests are routed for review on this turn and subsequent turns.",
     "anyOf": [
       {
         "$ref": "#/definitions/ApprovalsReviewer"
       },
       {
         "type": "null"
       }
     ]
   },
   "clientUserMessageId": {
     "type": [
       "string",
       "null"
     ]
   },
   "collaborationMode": {
     "description": "EXPERIMENTAL - Set a pre-set collaboration mode. Takes precedence over model, reasoning_effort, and developer instructions if set.\n\nFor `collaboration_mode.settings.developer_instructions`, `null` means \"use the built-in instructions for the selected mode\".",
     "anyOf": [
       {
         "$ref": "#/definitions/CollaborationMode"
       },
       {
         "type": "null"
       }
     ]
   },
   "cwd": {
     "description": "Override the working directory for this turn and subsequent turns.",
     "type": [
       "string",
       "null"
     ]
   },
   "effort": {
     "description": "Override the reasoning effort for this turn and subsequent turns.",
     "anyOf": [
       {
         "$ref": "#/definitions/ReasoningEffort"
       },
       {
         "type": "null"
       }
     ]
   },
   "environments": {
     "description": "Optional environments for this turn and subsequent turns.\n\nOmitted uses the thread sticky environments. Empty disables environment access for this turn. Non-empty selects the first environment as the current turn environment for this turn.",
     "type": [
       "array",
       "null"
     ],
     "items": {
       "$ref": "#/definitions/TurnEnvironmentParams"
     }
   },
   "input": {
     "type": "array",
     "items": {
       "$ref": "#/definitions/UserInput"
     }
   },
   "model": {
     "description": "Override the model for this turn and subsequent turns.",
     "type": [
       "string",
       "null"
     ]
   },
   "multiAgentMode": {
     "description": "@deprecated Ignored. Use `effort: \"ultra\"` for proactive multi-agent behavior.",
     "anyOf": [
       {
         "$ref": "#/definitions/MultiAgentMode"
       },
       {
         "type": "null"
       }
     ]
   },
   "outputSchema": {
     "description": "Optional JSON Schema used to constrain the final assistant message for this turn."
   },
   "permissions": {
     "description": "Select a named permissions profile id for this turn and subsequent turns. Cannot be combined with `sandboxPolicy`.",
     "type": [
       "string",
       "null"
     ]
   },
   "personality": {
     "description": "Override the personality for this turn and subsequent turns.",
     "anyOf": [
       {
         "$ref": "#/definitions/Personality"
       },
       {
         "type": "null"
       }
     ]
   },
   "responsesapiClientMetadata": {
     "description": "Optional metadata to enrich Codex's ResponsesAPI turn metadata.\n\nEntries are flattened into the JSON string sent as `client_metadata[\"x-codex-turn-metadata\"]` on ResponsesAPI HTTP and websocket requests.\n\nThey are not sent as top-level ResponsesAPI `client_metadata` keys, and reserved keys such as `session_id`, `thread_id`, `turn_id`, and `window_id` cannot be overridden.",
     "type": [
       "object",
       "null"
     ],
     "additionalProperties": {
       "type": "string"
     }
   },
   "runtimeWorkspaceRoots": {
     "description": "Replace the thread's runtime workspace roots for this turn and subsequent turns. Paths must be absolute.",
     "type": [
       "array",
       "null"
     ],
     "items": {
       "$ref": "#/definitions/AbsolutePathBuf"
     }
   },
   "sandboxPolicy": {
     "description": "Override the sandbox policy for this turn and subsequent turns.",
     "anyOf": [
       {
         "$ref": "#/definitions/SandboxPolicy"
       },
       {
         "type": "null"
       }
     ]
   },
   "serviceTier": {
     "description": "Override the service tier for this turn and subsequent turns.",
     "type": [
       "string",
       "null"
     ]
   },
   "summary": {
     "description": "Override the reasoning summary for this turn and subsequent turns.",
     "anyOf": [
       {
         "$ref": "#/definitions/ReasoningSummary"
       },
       {
         "type": "null"
       }
     ]
   },
   "threadId": {
     "type": "string"
   }
 },
 "$schema": "http://json-schema.org/draft-07/schema#"
}

Fields

additional_context: Option<HashMap<String, AdditionalContextEntry>>

Optional client-provided context fragments keyed by an opaque source identifier.

approval_policy: Option<AskForApproval>

Override the approval policy for this turn and subsequent turns.

approvals_reviewer: Option<ApprovalsReviewer>

Override where approval requests are routed for review on this turn and subsequent turns.

client_user_message_id: Option<String>

collaboration_mode: Option<CollaborationMode>

EXPERIMENTAL - Set a pre-set collaboration mode. Takes precedence over model, reasoning_effort, and developer instructions if set.

For collaboration_mode.settings.developer_instructions, null means “use the built-in instructions for the selected mode”.

cwd: Option<String>

Override the working directory for this turn and subsequent turns.

effort: Option<ReasoningEffort>

Override the reasoning effort for this turn and subsequent turns.

environments: Option<Vec<TurnEnvironmentParams>>

Optional environments for this turn and subsequent turns.

Omitted uses the thread sticky environments. Empty disables environment access for this turn. Non-empty selects the first environment as the current turn environment for this turn.

input: Vec<UserInput>

model: Option<String>

Override the model for this turn and subsequent turns.

multi_agent_mode: Option<MultiAgentMode>

@deprecated Ignored. Use effort: "ultra" for proactive multi-agent behavior.

output_schema: Option<Value>

Optional JSON Schema used to constrain the final assistant message for this turn.

permissions: Option<String>

Select a named permissions profile id for this turn and subsequent turns. Cannot be combined with sandboxPolicy.

personality: Option<Personality>

Override the personality for this turn and subsequent turns.

responsesapi_client_metadata: Option<HashMap<String, String>>

Optional metadata to enrich Codex’s ResponsesAPI turn metadata.

Entries are flattened into the JSON string sent as client_metadata["x-codex-turn-metadata"] on ResponsesAPI HTTP and websocket requests.

They are not sent as top-level ResponsesAPI client_metadata keys, and reserved keys such as session_id, thread_id, turn_id, and window_id cannot be overridden.

runtime_workspace_roots: Option<Vec<AbsolutePathBuf>>

Replace the thread’s runtime workspace roots for this turn and subsequent turns. Paths must be absolute.

sandbox_policy: Option<SandboxPolicy>

Override the sandbox policy for this turn and subsequent turns.

service_tier: Option<String>

Override the service tier for this turn and subsequent turns.

summary: Option<ReasoningSummary>

Override the reasoning summary for this turn and subsequent turns.

thread_id: String

Implementations

impl TurnStartParams

pub fn text(thread_id: impl Into<String>, text: impl Into<String>) -> Self

pub fn model(self, model: impl Into<String>) -> Self

pub fn cwd(self, cwd: impl Into<String>) -> Self

Trait Implementations

impl Clone for TurnStartParams

fn clone(&self) -> TurnStartParams

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for TurnStartParams

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for TurnStartParams

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for TurnStartParams

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.