Struct DenyAllToolPolicy 

pub struct DenyAllToolPolicy;

A ToolPolicy that requires approval for every tool call.

Combined with a deny-all approval handler, this denies every tool for the agent that carries it. Used by multi-agent runtimes as a fallback when a child should have “no permission” but its parent has no ToolPolicy of its own to inherit (so there is no notion of which tools are “dangerous”).

Trait Implementations

impl Clone for DenyAllToolPolicy

fn clone(&self) -> DenyAllToolPolicy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DenyAllToolPolicy

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

Formats the value using the given formatter.

impl Default for DenyAllToolPolicy

fn default() -> DenyAllToolPolicy

Returns the “default value” for a type.

impl ToolPolicy for DenyAllToolPolicy

fn evaluate_approval<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, tool_name: &'life1 str, _args: &'life2 Value, ) -> Pin<Box<dyn Future<Output = Option<ApprovalRequest>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Called before every tool call.

fn before_call( &self, tool_name: &str, args: &Value, ctx: &ToolContext, ) -> AgentResult<()>

Called immediately before a tool executes, after approval has been granted (or auto-approved).

fn after_call( &self, tool_name: &str, args: &Value, result: &[Content], ctx: &ToolContext, ) -> AgentResult<()>

Called immediately after a tool executes, before the result is returned to the LLM.