Struct MultiAgentDefensePipeline 

pub struct MultiAgentDefensePipeline { /* private fields */ }

Orchestrates multi-agent security: communication policy enforcement, privilege boundary checks, and message content scanning.

Implementations

impl MultiAgentDefensePipeline

pub fn new() -> Self

Create a pipeline with default configuration.

pub fn with_config(config: MultiAgentConfig) -> Self

Create a pipeline from explicit configuration.

pub fn register_agent(&mut self, profile: AgentProfile)

Register an agent profile for privilege and communication tracking.

pub fn allow_communication( &mut self, source: AgentId, target: AgentId, level: PermissionLevel, )

Grant communication permission between two agents.

pub fn deny_communication(&mut self, source: AgentId, target: AgentId)

Deny communication between two agents.

pub fn scan_message(&self, message: &InterAgentMessage) -> MessageScanResult

Scan message content for injection and exfiltration threats.

pub fn process_message(&mut self, message: InterAgentMessage) -> ProcessResult

Process a message through the full pipeline: permission check, privilege check (for delegations), and content scanning.

pub fn check_delegation( &self, from: &AgentId, to: &AgentId, tool: &str, ) -> DelegationCheck

Check delegation between two agents for a specific tool.

pub fn validate_message_flow( &self, source: &AgentId, target: &AgentId, ) -> FlowValidation

Validate whether a message is permitted to flow between two agents.

pub fn to_security_findings(result: &ProcessResult) -> Vec<SecurityFinding>

Convert a process result into security findings.

pub fn message_count(&self) -> usize

Number of messages currently in the log.

Trait Implementations

impl Debug for MultiAgentDefensePipeline

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

Formats the value using the given formatter.

impl Default for MultiAgentDefensePipeline

fn default() -> Self

Returns the “default value” for a type.