Struct AutonomyConfig 

pub struct AutonomyConfig {

    pub level: AutonomyLevel,
    pub workspace_only: bool,
    pub allowed_commands: Vec<String>,
    pub forbidden_paths: Vec<String>,
    pub max_actions_per_hour: u32,
    pub max_cost_per_day_cents: u32,
    pub require_approval_for_medium_risk: bool,
    pub block_high_risk_commands: bool,
    pub shell_env_passthrough: Vec<String>,
    pub auto_approve: Vec<String>,
    pub always_ask: Vec<String>,
    pub allowed_roots: Vec<String>,
    pub non_cli_excluded_tools: Vec<String>,
}

Autonomy and security policy configuration ([autonomy] section).

Controls what the agent is allowed to do: shell commands, filesystem access, risk approval gates, and per-policy budgets.

Fields

level: AutonomyLevel

Autonomy level: read_only, supervised (default), or full.

workspace_only: bool

Restrict absolute filesystem paths to workspace-relative references. Default: true. Resolved paths outside the workspace still require allowed_roots.

allowed_commands: Vec<String>

Allowlist of executable names permitted for shell execution.

forbidden_paths: Vec<String>

Explicit path denylist. Default includes system-critical paths and sensitive dotdirs.

max_actions_per_hour: u32

Maximum actions allowed per hour per policy. Default: 100.

max_cost_per_day_cents: u32

Maximum cost per day in cents per policy. Default: 1000.

require_approval_for_medium_risk: bool

Require explicit approval for medium-risk shell commands.

block_high_risk_commands: bool

Block high-risk shell commands even if allowlisted.

shell_env_passthrough: Vec<String>

Additional environment variables allowed for shell tool subprocesses.

These names are explicitly allowlisted and merged with the built-in safe baseline (PATH, HOME, etc.) after env_clear().

auto_approve: Vec<String>

Tools that never require approval (e.g. read-only tools).

always_ask: Vec<String>

Tools that always require interactive approval, even after “Always”.

allowed_roots: Vec<String>

Extra directory roots the agent may read/write outside the workspace. Supports absolute, ~/..., and workspace-relative entries. Resolved paths under any of these roots pass is_resolved_path_allowed.

non_cli_excluded_tools: Vec<String>

Tools to exclude from non-CLI channels (e.g. Telegram, Discord).

When a tool is listed here, non-CLI channels will not expose it to the model in tool specs.

Implementations

impl AutonomyConfig

pub fn ensure_default_auto_approve(&mut self)

Merge the built-in default auto_approve entries into the current list, preserving any user-supplied additions.

Trait Implementations

impl Clone for AutonomyConfig

fn clone(&self) -> AutonomyConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AutonomyConfig

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

Formats the value using the given formatter.

impl Default for AutonomyConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for AutonomyConfig

where AutonomyConfig: Default,

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for AutonomyConfig

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl Serialize for AutonomyConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.