Struct SessionContextV1 

pub struct SessionContextV1 {

    pub grant_id: SessionGrantId,
    pub grant_revision: HarnessRevision,
    pub actor_run: CallerRunV1,
    pub task: Option<RedactedTaskV1>,
    pub sibling_runs: Vec<RedactedRunV1>,
    pub read_permissions: HarnessReadPermissionsV1,
    pub monitoring_visibility: HarnessMonitoringVisibilityV1,
    pub child_task_count: u64,
    pub child_task_subtree_depth: u64,
    pub task_create: bool,
    pub task_mutate: bool,
    pub allowed_tool_ids: Vec<String>,
    pub history_message_count: Option<u64>,
    pub completed_turn_count: Option<u64>,
    pub total_tokens: Option<u64>,
}

Fields

grant_id: SessionGrantId

grant_revision: HarnessRevision

actor_run: CallerRunV1

task: Option<RedactedTaskV1>

The calling run’s own task, redacted the same way as TaskGet – present only when the grant’s task scope makes it visible (mirrors actor_run.task_id).

sibling_runs: Vec<RedactedRunV1>

Every OTHER run recorded against task (“what previous sessions did on it”), redacted the same way as RunGet. Gated solely by task being visible – independent of the grant’s runs read scope, which governs the general run-browsing tools instead. Always empty when task is None; never includes runs of another task.

read_permissions: HarnessReadPermissionsV1

monitoring_visibility: HarnessMonitoringVisibilityV1

child_task_count: u64

Observation, not a budget: the number of tasks parented directly under the calling run’s own task right now (HarnessEngine::task_child_count), 0 if it has no children. Unconditional – computed from actor_run’s own task the same way the retired maximum_child_count wire field it replaces was (independent of whether task above is redacted to None). That retired field stated a cap that was validated for range but compared against no actual child count anywhere in the service – a session could read a limit here and still create children past it. The owner’s ruling: do not cap a session’s children, detect them. Nothing in this wire enforces an upper bound on this count.

child_task_subtree_depth: u64

Observation, not a budget: the depth of the deepest chain of tasks parented (transitively) under the calling run’s own task right now (HarnessEngine::task_subtree_depth) – 0 if it has no children, 1 if it has children but no grandchildren, and so on. Unconditional in the same sense as child_task_count above. Replaces the retired maximum_child_depth wire field for the same reason child_task_ count replaces maximum_child_count (see that field’s doc comment). Nothing in this wire enforces an upper bound on this depth.

task_create: bool

D5, Slice D: mirrors grant.task_permissions.create – gates g4a_task_create in allowed_tool_ids.

task_mutate: bool

D5, Slice D: mirrors grant.task_permissions.mutate – gates g4a_task_move in allowed_tool_ids.

allowed_tool_ids: Vec<String>

history_message_count: Option<u64>

completed_turn_count: Option<u64>

total_tokens: Option<u64>

Implementations

impl SessionContextV1

pub fn validate(&self) -> Result<(), HarnessReadApiError>

Trait Implementations

impl Clone for SessionContextV1

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SessionContextV1

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for SessionContextV1

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for SessionContextV1

impl PartialEq for SessionContextV1

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for SessionContextV1

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 SessionContextV1