Struct TrustGateExecutor 

pub struct TrustGateExecutor<T: ToolExecutor> { /* private fields */ }

Wraps an inner ToolExecutor and applies trust-level permission overlays.

Implementations

impl<T: ToolExecutor> TrustGateExecutor<T>

pub fn new(inner: T, policy: PermissionPolicy) -> Self

pub fn mcp_tool_ids_handle(&self) -> Arc<RwLock<HashSet<String>>>

Returns the shared MCP tool ID set so the caller can populate it after MCP servers have connected (and after TrustGateExecutor has been wrapped in a DynExecutor).

pub fn with_trust_floor(self, floor: TurnTrustFloor) -> Self

Replaces this gate’s trust floor with an externally-owned, shared one (#6701).

Use when another component (e.g. SkillTrustGate) must observe and fold the exact same cell this gate reads in check_trust — pass the same TurnTrustFloor to both instead of relying on trust_floor, which can only be called after this gate already exists.

pub fn trust_floor(&self) -> TurnTrustFloor

Returns a clone of the shared per-turn trust floor (#6701).

Cloning is cheap (an Arc clone) and shares the same underlying cell — callers that need to downgrade trust from outside the ToolExecutor trait chain (e.g. a subagent spawn applying an inherited trust cap, or SkillTrustGate::resolve_body degrading on a Quarantined body read) can call TurnTrustFloor::fold on the returned handle directly instead of routing a set_effective_trust call back down through every wrapping executor layer.

pub fn set_effective_trust(&self, level: SkillTrustLevel)

pub fn effective_trust(&self) -> SkillTrustLevel

Trait Implementations

impl<T: ToolExecutor + Debug> Debug for TrustGateExecutor<T>

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

Formats the value using the given formatter.

impl<T: ToolExecutor> ToolExecutor for TrustGateExecutor<T>

fn requires_confirmation(&self, call: &ToolCall) -> bool

Returns true when the current policy would require confirmation for call.

Mirrors the decision in execute_tool_call without executing the tool. The speculative engine calls this to skip dispatch for tools that require user approval.

async fn execute(&self, response: &str) -> Result<Option<ToolOutput>, ToolError>

Parse response for fenced tool blocks and execute them.

async fn execute_confirmed( &self, response: &str, ) -> Result<Option<ToolOutput>, ToolError>

Execute bypassing confirmation checks (called after user approves).

fn tool_definitions(&self) -> Vec<ToolDef>

Return the tool definitions this executor can handle.

async fn execute_tool_call( &self, call: &ToolCall, ) -> Result<Option<ToolOutput>, ToolError>

Execute a structured tool call. Returns Ok(None) if call.tool_id is not handled.

async fn execute_tool_call_confirmed( &self, call: &ToolCall, ) -> Result<Option<ToolOutput>, ToolError>

Execute a structured tool call bypassing confirmation checks.

fn set_skill_env(&self, env: Option<HashMap<String, String>>)

Inject environment variables for the currently active skill. No-op by default.

fn is_tool_retryable(&self, tool_id: &str) -> bool

Whether the executor can safely retry this tool call on a transient error.

fn is_tool_speculatable(&self, tool_id: &str) -> bool

Whether a tool call can be safely dispatched speculatively (before the LLM finishes).

fn checkpoint_undo(&self, n: usize) -> CheckpointActionResult

Undo the last n checkpointed write commands.

fn checkpoint_redo(&self) -> CheckpointActionResult

Redo the last undone checkpoint.

fn checkpoint_list(&self) -> CheckpointListResult

List the current undo stack entries and redo depth.

fn set_effective_trust(&self, level: SkillTrustLevel)

Set the effective trust level for the currently active skill. No-op by default.