Struct SecurityConfig 

pub struct SecurityConfig {
    pub bash_risk_threshold: String,
    pub sandbox: String,
    pub sandbox_network: bool,
    pub sandbox_fallback: String,
    pub network_allowlist: Vec<String>,
    pub network_denylist: Vec<String>,
    pub redact_secrets: bool,
}

Settings for shell safety: risk classification, OS sandbox, and network domain policy for HTTP tools.

The risk classifier inspects the command string a model asked to run. The sandbox (when enabled) confines the process that runs it. They stack; neither replaces the other. See the README security section.

network_allowlist / network_denylist gate webfetch, websearch, and GitHub API tools by host pattern. Shell network stays binary (sandbox_network); domain filtering does not apply inside the shell.

Fields

bash_risk_threshold: String

Lowest risk level that requires confirmation: caution, destructive or off. Commands classified catastrophic are always refused and this setting cannot override that.

sandbox: String

OS sandbox for bash / shell: off | workspace (default).

sandbox_network: bool

When sandbox = "workspace", whether the sandboxed shell may use the network. Default true so cargo / npm / git keep working.

sandbox_fallback: String

When the requested sandbox cannot be applied (no bwrap, non-Linux): allow (default — warn and run on host) or deny (fail the tool call).

network_allowlist: Vec<String>

Host patterns allowed for outbound HTTP tools. Empty (default) means unrestricted. See NetworkPolicy.

network_denylist: Vec<String>

Host patterns always blocked for outbound HTTP tools (wins over allowlist).

redact_secrets: bool

Replace credential-looking values (environment variables named like secrets, known token shapes, URL passwords) with placeholders in every request sent to a provider, and restore them inside the model’s tool arguments before execution. Default true.

Implementations

impl SecurityConfig

pub fn network_policy(&self) -> NetworkPolicy

Build the HTTP-tool network policy from config lists.

pub fn sandbox_settings(&self) -> SandboxSettings

Resolved OS sandbox policy for shell tools.

Trait Implementations

impl Clone for SecurityConfig

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 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

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

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.