agentcarousel 0.8.1

Unit tests for AI agents. Run behavioral tests in CI, score with an LLM judge, and export signed evidence your auditors accept.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

//! Human-readable and machine-readable **output**: terminal tables, JSON, and persisted
//! history (SQLite).

#[cfg(feature = "dashboard")]
pub mod dashboard;
mod history;
mod terminal;

pub use history::{
    fetch_run, find_previous_run, find_tagged_run, list_full_runs, list_full_runs_by_skill,
    list_runs, persist_run, tag_run, HistoryError, RunListing,
};
pub use terminal::{print_audit, print_terminal, print_terminal_summary};

pub fn print_json(run: &agentcarousel_core::Run) {
    let payload = serde_json::to_string_pretty(run).unwrap_or_else(|_| "{}".to_string());
    println!("{payload}");
}