Struct RunConfig 

pub struct RunConfig {
    pub schema: SchemaConfig,
    pub predicate: Option<ToolPredicate>,
    pub tool_filter: Option<ToolNamePredicate>,
    pub assertions: AssertionSet,
    pub in_band_error_forbidden: bool,
    pub state_machine: StateMachineConfig,
    pub pre_run_hook: Option<PreRunHook>,
    pub full_trace: bool,
    pub show_uncallable: bool,
    pub trace_sink: Option<Arc<dyn TraceSink>>,
    pub lints: LintSuite,
    /* private fields */
}

Top-level configuration for executing a tooltest run.

Downstream crates cannot construct this type via a struct literal; use RunConfig::new and the builder methods to ensure invariants are validated.

use tooltest_core::RunConfig;

let _ = RunConfig {
    uncallable_limit: 0,
    ..RunConfig::new()
};

Fields

schema: SchemaConfig

MCP schema configuration.

predicate: Option<ToolPredicate>

Optional predicate to filter eligible tools.

tool_filter: Option<ToolNamePredicate>

Optional predicate to filter eligible tools by name.

assertions: AssertionSet

Assertion rules to evaluate during the run.

in_band_error_forbidden: bool

Whether tool result error responses (isError) should fail the run.

state_machine: StateMachineConfig

State-machine generator configuration.

pre_run_hook: Option<PreRunHook>

Optional pre-run hook to execute before validation and each case.

full_trace: bool

Emit full tool responses in traces instead of compact invocation-only entries.

show_uncallable: bool

Include uncallable tool traces when coverage validation fails.

trace_sink: Option<Arc<dyn TraceSink>>

Optional trace sink for streaming per-case traces.

lints: LintSuite

Configured lint rules for the run.

Implementations

impl RunConfig

pub fn new() -> Self

Creates a run configuration with defaults for schema and assertions.

The state-machine generator is always used, and it is strict by default (required values must come from the corpus unless lenient sourcing is enabled).

pub fn with_schema(self, schema: SchemaConfig) -> Self

Sets the schema configuration.

pub fn with_predicate(self, predicate: ToolPredicate) -> Self

Sets the tool predicate used for eligibility filtering.

pub fn with_tool_filter(self, predicate: ToolNamePredicate) -> Self

Sets the tool name predicate used for eligibility filtering.

pub fn with_assertions(self, assertions: AssertionSet) -> Self

Sets the assertion rules for the run.

pub fn with_in_band_error_forbidden(self, forbidden: bool) -> Self

Sets whether tool result error responses (isError) should fail the run.

pub fn with_state_machine(self, state_machine: StateMachineConfig) -> Self

Sets the state-machine generator configuration.

pub fn with_pre_run_hook(self, hook: PreRunHook) -> Self

Sets the pre-run hook for this run.

pub fn with_full_trace(self, enabled: bool) -> Self

Enables full trace output with tool responses.

pub fn with_show_uncallable(self, enabled: bool) -> Self

Enables uncallable tool trace output for coverage validation failures.

pub fn with_uncallable_limit(self, limit: usize) -> Result<Self, String>

Sets the call limit for uncallable tool traces.

pub fn uncallable_limit(&self) -> usize

Returns the call limit for uncallable tool traces.

pub fn with_trace_sink(self, sink: Arc<dyn TraceSink>) -> Self

Sets a trace sink that receives per-case traces.

pub fn with_lints(self, lints: LintSuite) -> Self

Sets the configured lints for this run.

Trait Implementations

impl Clone for RunConfig

fn clone(&self) -> RunConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RunConfig

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

Formats the value using the given formatter.

impl Default for RunConfig

fn default() -> Self

Returns the “default value” for a type.