Struct SpawnArgs 

pub struct SpawnArgs {

    pub run_id: String,
    pub blueprint_path: String,
    pub task: String,
    pub regions: HashMap<String, String>,
    pub model: Option<String>,
    pub workdir: String,
    pub metadata: HashMap<String, String>,
    pub callback_url: Option<String>,
    pub callback_secret: Option<String>,
    pub yolo: bool,
    pub no_seed_commands: bool,
    pub allow: Vec<String>,
    pub max_depth: Option<usize>,
    pub parent_run_id: Option<String>,
}

The parameters for spawning an agent into the world. The runtime doesn’t know how to load blueprints or resolve tools - that policy lives in the Spawner the daemon installs - so this just carries the raw request.

Fields

run_id: String

The run id to give the new agent (its directory / control key).

blueprint_path: String

Path to the agent manifest directory or bundle.

task: String

The task prompt. Seeded into the region keyed task (see crate::context_setup::init_window_seeded); a matching regions entry, if present, overrides it.

regions: HashMap<String, String>

Literal seed content for named caller-input regions, keyed by the region’s caller-input name. Merged over task at spawn. #[serde(default)] keeps older requests (which never sent this) deserializing to an empty map.

model: Option<String>

Optional model override (provider/model or model).

workdir: String

Working directory for tool execution.

metadata: HashMap<String, String>

Custom key/value metadata from the request.

callback_url: Option<String>

Webhook to POST on completion/error (surfaced in the run metadata).

callback_secret: Option<String>

Optional shared secret for HMAC-SHA256 signing the webhook body.

yolo: bool

Run this agent unattended (the --yolo launch override): approve every tool call, waive the taint gate, and auto-answer the agent’s own prompts (ask_user_*, blueprint interaction points) rather than parking on the interaction hub for a person who isn’t there.

no_seed_commands: bool

Refuse this run’s seed = { command = ... } regions (the --no-seed-commands launch override). Command seeds execute at spawn, before any approval prompt, so this is the per-run counterpart to the [security] allow_seed_commands config switch.

allow: Vec<String>

Tools to allow outright for this run (the --allow launch override).

max_depth: Option<usize>

Override the blueprint’s max sub-agent tree depth.

parent_run_id: Option<String>

The run id of this agent’s parent, when it is a sub-agent / fan-out worker. Persisted in the run metadata so observers (dashboard, serve tree) can nest children under their parent. None for a top-level run.

Trait Implementations

impl Clone for SpawnArgs

fn clone(&self) -> SpawnArgs

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SpawnArgs

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

Formats the value using the given formatter.

impl Default for SpawnArgs

fn default() -> SpawnArgs

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for SpawnArgs

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for SpawnArgs

fn eq(&self, other: &SpawnArgs) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for SpawnArgs

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for SpawnArgs