Struct AskArgs 

pub struct AskArgs {

    pub targets: Vec<String>,
    pub question: Option<String>,
    pub uri: Option<String>,
    pub scope: Option<String>,
    pub top_k: usize,
    pub snippet_chars: usize,
    pub cursor: Option<String>,
    pub mode: AskModeArg,
    pub json: bool,
    pub context_only: bool,
    pub sources: bool,
    pub mask_pii: bool,
    pub memories: bool,
    pub llm_context_depth: Option<usize>,
    pub start: Option<String>,
    pub end: Option<String>,
    pub use_model: Option<String>,
    pub query_embedding_model: Option<String>,
    pub as_of_frame: Option<u64>,
    pub as_of_ts: Option<i64>,
    pub system_prompt: Option<String>,
    pub no_rerank: bool,
    pub adaptive: bool,
    pub min_relevancy: f32,
    pub max_k: usize,
    pub adaptive_strategy: AdaptiveStrategyArg,
}

Arguments for the ask subcommand

Fields

targets: Vec<String>

question: Option<String>

uri: Option<String>

scope: Option<String>

top_k: usize

snippet_chars: usize

cursor: Option<String>

mode: AskModeArg

json: bool

context_only: bool

sources: bool

Show detailed source information for each citation

mask_pii: bool

Mask PII (emails, SSNs, phone numbers, etc.) in context before sending to LLM

memories: bool

Include structured memory cards in the context (facts, preferences, etc.)

llm_context_depth: Option<usize>

Maximum characters of retrieval context to send to remote LLMs (overrides MEMVID_LLM_CONTEXT_BUDGET)

start: Option<String>

end: Option<String>

use_model: Option<String>

query_embedding_model: Option<String>

Embedding model to use for query (must match the model used during ingestion) Options: bge-small, bge-base, nomic, gte-large, openai, openai-small, openai-ada

as_of_frame: Option<u64>

Replay: Filter to frames with ID <= AS_OF_FRAME (time-travel view)

as_of_ts: Option<i64>

Replay: Filter to frames with timestamp <= AS_OF_TS (time-travel view)

system_prompt: Option<String>

Override the default system prompt (useful for providing date context like “Today is March 27, 2023”)

no_rerank: bool

Skip cross-encoder reranking (useful in gated environments where model downloads are blocked)

adaptive: bool

Enable adaptive retrieval: dynamically determine how many results to return based on relevancy score distribution. Instead of fixed top-k, pulls all results above a relevancy threshold. Use this when answers may span multiple chunks.

min_relevancy: f32

Minimum relevancy ratio vs top score (0.0-1.0). Results below this threshold are excluded. Example: 0.5 means only include results with score >= 50% of the top result’s score. Only used when –adaptive is enabled.

max_k: usize

Maximum results to consider for adaptive retrieval (over-retrieval limit). Set high enough to capture all potentially relevant results. Only used when –adaptive is enabled.

adaptive_strategy: AdaptiveStrategyArg

Adaptive cutoff strategy: relative (default), absolute, cliff, elbow, or combined

Trait Implementations

impl Args for AskArgs

fn group_id() -> Option<Id>

Report the ArgGroup::id for this set of arguments

fn augment_args<'b>(__clap_app: Command) -> Command

Append to Command so it can instantiate Self via FromArgMatches::from_arg_matches_mut

fn augment_args_for_update<'b>(__clap_app: Command) -> Command

Append to Command so it can instantiate self via FromArgMatches::update_from_arg_matches_mut

impl FromArgMatches for AskArgs

fn from_arg_matches(__clap_arg_matches: &ArgMatches) -> Result<Self, Error>

Instantiate Self from ArgMatches, parsing the arguments as needed.

fn from_arg_matches_mut( __clap_arg_matches: &mut ArgMatches, ) -> Result<Self, Error>

Instantiate Self from ArgMatches, parsing the arguments as needed.

fn update_from_arg_matches( &mut self, __clap_arg_matches: &ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

fn update_from_arg_matches_mut( &mut self, __clap_arg_matches: &mut ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.