Struct SecurityConfig 

pub struct SecurityConfig {
    pub mode: SecurityMode,
    pub spotlight_external: bool,
    pub trifecta_gate: bool,
    pub pin_mcp_schemas: bool,
    pub gate_secret_reads: bool,
    pub detect_injection: bool,
    pub guard_threshold_percent: u8,
    pub guard_model: String,
    pub trusted_mcp_servers: Vec<String>,
}

[security] configuration. The runtime substrate lives in crate::security; this is the typed, layerable shape hosts persist and merge. Defaults are on (deterministic, free) so the runtime is secure-by-default; the trifecta gate only takes effect where an interactive approval policy is installed.

Fields

mode: SecurityMode

spotlight_external: bool

Frame untrusted external tool/MCP output in spotlight delimiters.

trifecta_gate: bool

Apply the lethal-trifecta gate: force confirmation when tainted context reaches an exfiltration-capable or destructive tool.

pin_mcp_schemas: bool

Pin + hash MCP tool schemas; require re-approval when a server mutates a tool description after first approval (rug-pull defense).

gate_secret_reads: bool

Also gate reads of well-known secret/credential files while tainted.

detect_injection: bool

Score untrusted content with an injection classifier (Layer 2). Implied by mode = "local-ml"; can be opted into under spotlight/strict too. The classifier is the built-in heuristic unless a harn-guard neural model is registered. A flagged score tightens the trifecta gate.

guard_threshold_percent: u8

Malicious-probability threshold, as a percent in [0, 100], at or above which the classifier marks content as flagged. Kept as an integer so the config stays Eq-comparable and round-trips losslessly.

guard_model: String

Selector for the downloadable neural classifier used when detect_injection is on: a harn guard catalog name (the default) or a path to a model directory. Resolved lazily by the host’s harn-guard loader; an empty value or an uninstalled model keeps the built-in heuristic. Ignored by binaries built without the guard inference backend.

trusted_mcp_servers: Vec<String>

MCP servers the operator has explicitly trusted (skip taint + pinning).

Trait Implementations

impl Clone for SecurityConfig

fn clone(&self) -> SecurityConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SecurityConfig

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

Formats the value using the given formatter.

impl Default for SecurityConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for SecurityConfig

where SecurityConfig: Default,

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for SecurityConfig

impl PartialEq for SecurityConfig

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for SecurityConfig

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 SecurityConfig