/**
 * dsh-eval — agent evaluation over headless dsh profiles.
 *
 * The eval bundle mounts this command-line plugin over dsh-base. It parses
 * `dsh --profile eval run <benchmark.yaml>` / `report <run.json>`, spawns one
 * headless dsh subprocess per case x trial, harvests each persisted session
 * log as a trace, folds automatic metrics, scripted grading (task success,
 * tool-selection accuracy), and cross-run comparison, and writes a JSON run
 * report. LLM-judge metrics (final-answer score, hallucination) and paired
 * same-seed trials are deferred to later milestones.
 *
 * @module dsh-eval
 */
import type { Context } from '@deepseek-ai/cordis';
import { type EvalStartupValues } from './command.ts';
import { type JudgeChat } from './judge.ts';
/** Stable Cordis plugin name. */
export declare const name = "eval";
/** The launcher-provided command line is the only required service. */
export declare const inject: string[];
/** Process-facing output streams; tests substitute captures. */
export interface EvalIo {
    stdout: {
        write(chunk: string): unknown;
    };
    stderr: {
        write(chunk: string): unknown;
    };
}
/** The process streams the eval app writes to; tests substitute captures. */
export declare const internals: EvalIo;
/**
 * Resolve the production judge chat seam from the mounted llm service, or
 * undefined when the composition carries no llm service.
 * @param ctx - the plugin context.
 * @returns a chat seam over the llm stream, or undefined without llm.
 */
export declare function resolveJudgeChat(ctx: Context): JudgeChat | undefined;
/**
 * Execute one eval invocation to completion.
 * @param values - the resolved command-line invocation.
 * @param io - output streams for the report and diagnostics.
 * @param services - injected runtime seams, e.g. the LLM-judge chat.
 * @returns the process exit code; nonzero for any failure or a run with no
 *   completed trials.
 */
export declare function executeEval(values: EvalStartupValues, io: EvalIo, services?: {
    judgeChat?: JudgeChat | undefined;
}): Promise<number>;
/**
 * Mount the eval command-line app: parse the invocation, run it asynchronously,
 * and request launcher exit with the result code.
 * @param ctx - plugin context carrying the command line and launcher exit.
 */
export declare function apply(ctx: Context): void;