Struct SemanticCacheConfig 

pub struct SemanticCacheConfig {
    pub enabled: bool,
    pub embedding_url: String,
    pub embedding_model: String,
    pub embedding_dim: usize,
    pub threshold: f64,
    pub ttl_secs: u64,
    pub max_entries: usize,
    pub embedding_timeout_ms: u64,
    pub embedding_api_key: Option<String>,
}

Semantic cache config (DW-083 ai.semantic_cache): an embedding-similarity cache for AI prompts. A paraphrased prompt within the cosine-similarity threshold returns the cached response with no provider call and no token spend. Uses an external embedding service (OpenAI-compatible /v1/embeddings API) to vectorize prompts and hnsw_rs (pure Rust HNSW) for approximate nearest neighbor search. Compiled into every build (hnsw_rs is an unconditional dependency).

Fields

enabled: bool

Whether the semantic cache is enabled. Default false: an ai.semantic_cache block without enabled: true caches nothing (the engine compiles but stays inert).

embedding_url: String

The URL of the external embedding service (OpenAI-compatible /v1/embeddings API). Required when enabled. Validation rejects an empty or non-http(s) URL.

embedding_model: String

The model name passed to the embedding service in the model field of the POST body.

embedding_dim: usize

The dimension of the embedding vectors. Must match what the embedding service returns for the configured model; validation rejects 0.

threshold: f64

Cosine similarity threshold (0.0 to 1.0, inclusive). A cached entry is returned only if its cosine similarity to the query embedding is >= this threshold. Higher = stricter matching (fewer hits, more accurate). Default 0.85.

ttl_secs: u64

TTL for cached entries in seconds. Entries older than this are considered stale and not returned. Default 3600 (1 hour).

max_entries: usize

Maximum number of entries to cache. When the cache is full, it is reset (all entries evicted, the HNSW index rebuilt). Default 10 000.

embedding_timeout_ms: u64

Timeout for the embedding service HTTP call in milliseconds. Default 5000 (5 seconds).

embedding_api_key: Option<String>

Optional API key for the embedding service (sent as Authorization: Bearer <key>). Can be a ${...} secret reference (resolved at config-compile time). Inline values are redacted in config echoes; never logged.

Trait Implementations

impl Clone for SemanticCacheConfig

fn clone(&self) -> SemanticCacheConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SemanticCacheConfig

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for SemanticCacheConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for SemanticCacheConfig

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 PartialEq for SemanticCacheConfig

fn eq(&self, other: &SemanticCacheConfig) -> 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 SemanticCacheConfig

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 SemanticCacheConfig