ContextProcessingReport
SUPPORT · interface · 类型合同
Facts recorded for one processing run.
导入
ts
import type { ContextProcessingReport } from '@jangts/agent-core/support';js
/** @typedef {import('@jangts/agent-core/support').ContextProcessingReport} ContextProcessingReport */其他公开入口：@jangts/agent-core/support、@jangts/agent-core/support/context。
定义
ts
interface ContextProcessingReport字段结构
| 字段 | 类型 | 必需 | 说明 |
|---|---|---|---|
mode | "reduce" | "validate" | "audit" | 必需 | Policy mode that ran. |
selection | "auto" | "fixed" | "sequence" | "model" | undefined | 可省略 | Selection mode applied, when the policy reduced content. |
decisions | { mode: ContextSelection["mode"]; stepIds: string[]; reason: string; }[] | undefined | 可省略 | Per-step selection decisions, kept for auditing auto and model choices. |
beforeChars | number | 必需 | Total characters before and after processing. |
afterChars | number | 必需 | 见类型定义及调用场景。 |
tokens | { counterId: string; before: number; after: number; inputLimit: number; } | undefined | 可省略 | Token measurement, present only when a token budget was supplied. |
changes | { id: string; kind: "trim" | "compress" | "app"; fragmentId: string; beforeChars: number; afterChars: number; }[] | 必需 | Applied changes in execution order; app marks an App-supplied mutation. |
使用示例
- 上下文策略：验证、缩减与审计：Support 给出可替换的策略实现，Core 保留最终输入合同和预算检查。
显式使用只校验策略
入门 · Node.js
ts
import { Support } from '@jangts/agent-core';
import type { ContextPolicy } from '@jangts/agent-core/support/context';
const policy: ContextPolicy = { mode: 'validate' };
Support.validateContextPolicy(policy, [], []);
console.log(policy.mode);js
import { Support } from '@jangts/agent-core';
const policy = { mode: 'validate' };
Support.validateContextPolicy(policy, [], []);
console.log(policy.mode);从真实业务片段配置缩减顺序
配置 · Node.js
这些数字是示例业务的显式选择，不是 SDK 推荐适用于所有模型的容量。
ts
import { Support } from '@jangts/agent-core';
import type { ContextPolicy } from '@jangts/agent-core/support/context';
const policy: ContextPolicy = { mode: 'reduce', triggerChars: 1800, targetInputChars: 1500,
selection: { mode: 'sequence' },
steps: [{ id: 'recent-history', kind: 'trim', fragmentId: 'history', field: 'messages',
grouping: { kind: 'user-turn' }, keepGroups: 2, targetChars: 1000 }],
};
Support.validateContextPolicy(policy, [], []);
console.log(policy.steps);js
import { Support } from '@jangts/agent-core';
const policy = { mode: 'reduce', triggerChars: 1800, targetInputChars: 1500,
selection: { mode: 'sequence' },
steps: [{ id: 'recent-history', kind: 'trim', fragmentId: 'history', field: 'messages',
grouping: { kind: 'user-turn' }, keepGroups: 2, targetChars: 1000 }],
};
Support.validateContextPolicy(policy, [], []);
console.log(policy.steps);