Struct RuntimeProfileConfig 

pub struct RuntimeProfileConfig {

    pub agentic: bool,
    pub max_tool_iterations: usize,
    pub max_execution_tree_iterations: Option<usize>,
    pub max_actions_per_hour: u32,
    pub max_cost_per_day_cents: u32,
    pub shell_timeout_secs: u64,
    pub max_delegation_depth: u32,
    pub delegation_timeout_secs: Option<u64>,
    pub agentic_timeout_secs: Option<u64>,
    pub max_history_messages: Option<usize>,
    pub max_context_tokens: Option<usize>,
    pub context_compact_ratio: Option<f64>,
    pub compact_context: Option<bool>,
    pub parallel_tools: Option<bool>,
    pub single_tool_rounds: Option<bool>,
    pub tool_dispatcher: Option<String>,
    pub tool_call_dedup_exempt: Vec<String>,
    pub deferred_builtin_tools: bool,
    pub max_system_prompt_chars: Option<usize>,
    pub max_tool_result_chars: Option<usize>,
    pub keep_tool_context_turns: Option<usize>,
    pub memory_recall_limit: Option<usize>,
    pub prompt_injection_mode: Option<SkillsPromptInjectionMode>,
    pub strict_tool_parsing: bool,
    pub thinking: ThinkingConfig,
    pub history_pruning: HistoryPrunerConfig,
    pub eval: EvalConfig,
    pub auto_classify: Option<AutoClassifyConfig>,
    pub effort_routing: Option<EffortRoutingConfig>,
    pub context_compression: ContextCompressionConfig,
    pub tool_receipts: ToolReceiptsConfig,
    pub tool_filter_groups: Vec<ToolFilterGroup>,
}

Named runtime/LLM execution profile ([runtime_profiles.<alias>]).

Reusable operational tuning: agentic mode, iteration caps, context budget, parallel dispatch, resource ceilings, recursion depth, and the budget knobs that SecurityPolicy enforces with subagent parent-subset discipline. Anything authorization-shaped (allowed commands/tools/paths, approval gates, sandbox) lives on [risk_profiles.<alias>]. Anything model-provider shaped (model, temperature, max_tokens, timeout_secs) lives on [providers.models.<type>.<alias>].

Fields

agentic: bool

Enable agentic (multi-turn tool-call loop) mode.

max_tool_iterations: usize

Maximum tool-call iterations in agentic mode. 0 inherits the global default.

max_execution_tree_iterations: Option<usize>

Maximum aggregate loop iterations for one execution tree. Omitted disables it.

max_actions_per_hour: u32

Maximum actions allowed per hour. 0 is a hard zero budget — the per-sender rate tracker treats a max of 0 as always exhausted (PerSenderTracker::is_exhausted), blocking every action. For an effectively-unlimited budget use a high value (e.g. u32::MAX), not 0. SecurityPolicy::ensure_no_escalation_beyond rejects subagents that try to raise this above the parent’s value.

max_cost_per_day_cents: u32

Maximum cost per day in cents. 0 inherits the global limit. Parent-subset enforced for subagents.

shell_timeout_secs: u64

Shell subprocess timeout in seconds. 0 inherits the global timeout. Parent-subset enforced for subagents.

max_delegation_depth: u32

Maximum delegation recursion depth. 0 inherits the default.

delegation_timeout_secs: Option<u64>

Delegate call timeout in seconds. None inherits global delegate timeout.

agentic_timeout_secs: Option<u64>

Agentic delegate run timeout in seconds. None inherits global.

max_history_messages: Option<usize>

History retention limit per session. Structured Agent and legacy loop sessions count complete turns; channel caches count message rows. None inherits the default of 50.

max_context_tokens: Option<usize>

Maximum estimated tokens before proactive history trimming. None preserves the legacy 32,000-token default when context_compact_ratio is unset. In ratio mode this remains an optional downward cap. Every positive effective value is capped by the selected model capacity. 0 disables proactive token-budget trimming.

context_compact_ratio: Option<f64>

Opt-in fraction of the selected model’s context window at which proactive history trimming triggers (e.g. 0.9 = trim at 90%). None preserves the legacy absolute-budget behavior. Values outside (0, 1] are treated as unset.

compact_context: Option<bool>

Use compact bootstrap (6000 chars / 2 RAG chunks). None inherits.

parallel_tools: Option<bool>

Enable parallel tool execution per iteration. None inherits.

single_tool_rounds: Option<bool>

Request one native tool call per provider response when supported. Default off.

tool_dispatcher: Option<String>

Tool dispatch strategy (e.g. "auto"). None inherits.

tool_call_dedup_exempt: Vec<String>

Tools exempt from within-turn dedup check.

deferred_builtin_tools: bool

Expose only basic workspace and memory tool schemas eagerly, with other admitted built-ins discoverable through tool_search. Default false preserves eager exposure. This never changes tool permissions; if tool_search is not admitted, built-in schemas remain eager.

max_system_prompt_chars: Option<usize>

Maximum characters for the assembled system prompt. None inherits.

max_tool_result_chars: Option<usize>

Maximum characters for a single tool result. None inherits.

keep_tool_context_turns: Option<usize>

Number of recent turns whose full tool context is preserved. None inherits.

memory_recall_limit: Option<usize>

Maximum memory entries injected per turn. None inherits global default (5). Set to 0 for unlimited.

prompt_injection_mode: Option<SkillsPromptInjectionMode>

How skills are injected into the system prompt for agents on this profile. None inherits the global [skills] prompt_injection_mode; compact inlines only compact skill metadata and registers the read_skill tool, while full inlines full skill instructions. Resolved through Config::effective_skills_prompt_mode, the single point both the system-prompt builder and the read_skill tool-registration gate consult so they always agree on the effective mode.

strict_tool_parsing: bool

thinking: ThinkingConfig

history_pruning: HistoryPrunerConfig

eval: EvalConfig

auto_classify: Option<AutoClassifyConfig>

effort_routing: Option<EffortRoutingConfig>

context_compression: ContextCompressionConfig

tool_receipts: ToolReceiptsConfig

tool_filter_groups: Vec<ToolFilterGroup>

Implementations

impl RuntimeProfileConfig

pub fn configurable_prefix() -> &'static str

Returns the #[prefix] value for this Configurable struct.

pub fn init_requires_explicit_config() -> bool

True when this struct has a required leaf field that a bare/ancestor-prefix init_defaults scaffold must not materialize (it would fill the field with its Rust Default, which prune_empty_leaves then strips on save, leaving a partial sub-table that fails strict reload). Explicitly targeting this section (or a descendant of it) still scaffolds regardless of this flag, so the section can be materialized for the operator to fill in.

pub fn secret_fields(&self) -> Vec<SecretFieldInfo>

Returns metadata about all #[secret] fields on this struct and nested children.

pub fn secret_field_terminals() -> Vec<&'static str>

pub fn encrypt_secrets(&mut self, store: &SecretStore) -> Result<()>

Encrypt all secret fields in place using the provided store.

pub fn decrypt_secrets(&mut self, store: &SecretStore) -> Result<()>

Decrypt all secret fields in place using the provided store.

pub fn set_secret(&mut self, name: &str, value: String) -> Result<()>

Set a secret field by its full dotted name, dispatching to nested children.

pub fn prop_fields(&self) -> Vec<PropFieldInfo>

Returns metadata about all property fields on this struct and nested children.

pub fn get_prop(&self, name: &str) -> Result<String>

Get a property value by its full dotted name, returning it as a display string.

pub fn set_prop(&mut self, name: &str, value_str: &str) -> Result<()>

Set a property value by its full dotted name, parsing from string.

pub fn prop_is_secret(name: &str) -> bool

Check if a property name refers to a secret field (static, no instance needed).

pub fn init_defaults(&mut self, prefix: Option<&str>) -> Vec<&'static str>

Instantiate None nested sections whose prefix matches. Returns the prefixes that were initialized.

pub fn map_key_sections() -> Vec<MapKeySection>

Enumerate every map-keyed (HashMap<String, T>) and list-shaped (Vec<T>) section discoverable from this Configurable’s tree. The dashboard / CLI consume this to surface “+ Add” affordances without hardcoding the section list.

pub fn nested_section_help(name: &str) -> Option<&'static str>

Help blurb for a #[nested] field on this struct, sourced from the field-level /// docstring. Returns None for unknown names so callers can fall through to a different lookup.

pub fn nested_section_group(name: &str) -> Option<&'static str>

pub fn get_map_keys(&self, section_path: &str) -> Option<Vec<String>>

Return the current alias keys at section_path, or None if the path doesn’t resolve to a map-keyed section in this tree.

pub fn nested_option_entries(&self) -> Vec<NestedOptionEntry>

pub fn create_map_key( &mut self, section_path: &str, map_key: &str, ) -> Result<bool, String>

pub fn delete_map_key( &mut self, section_path: &str, map_key: &str, ) -> Result<bool, String>

pub fn rename_map_key( &mut self, section_path: &str, map_key: &str, new_key: &str, ) -> Result<bool, String>

Trait Implementations

impl Clone for RuntimeProfileConfig

fn clone(&self) -> RuntimeProfileConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RuntimeProfileConfig

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

Formats the value using the given formatter.

impl Default for RuntimeProfileConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for RuntimeProfileConfig

where RuntimeProfileConfig: 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 RuntimeProfileConfig

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 MaskSecrets for RuntimeProfileConfig

fn mask_secrets(&mut self)

fn restore_secrets_from(&mut self, current: &Self)

impl Serialize for RuntimeProfileConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.