---
title: "Minimum Viable Harness"
type: concept
tags:
  - "harness-engineering"
  - "agentic-ai"
  - "quick-start"
sources:
  - "Your-AI-Isn-t-Stupid---It-Just-Needs-a-Better-Harness--Lychee-Technology-Engineering-Blog"
last_updated: 2026-04-20
---

## Overview
Minimum Viable Harness——Day 1 即可构建的 4 个核心组件，使 Agent 能够优雅失败，为后续增加智能赢得权利。

## The 4 Components

### 1. `state.json`
单一结构化文件跟踪任务状态——如果进程死亡，可以从上次停止的地方继续。
```json
{
  "task_id": "report-001",
  "steps": {
    "fetch_competitor_a": "COMPLETED",
    "fetch_competitor_b": "IN_PROGRESS",
    "compare": "PENDING"
  },
  "last_updated": "2026-04-20T10:30:00Z"
}
```

### 2. Retry Wrapper
每个工具调用包装 try/catch，至少一次自动重试 + 指数退避：
```python
def tool_call_with_retry(func, max_attempts=3):
    for attempt in range(max_attempts):
        try:
            return func()
        except (TimeoutError, APIError) as e:
            if attempt < max_attempts - 1:
                sleep(exponential_backoff(attempt))
            else:
                raise
```

### 3. Schema Validator
每个 LLM 输出在接收前必须通过 JSON Schema 验证。格式不符触发重试，不触发崩溃：
```python
def validate_and_retry(llm_output, schema):
    try:
        validated = jsonschema.validate(llm_output, schema)
        return validated
    except ValidationError:
        return retry_with_error_feedback(llm_output, schema)
```

### 4. Tool Output Truncation
每个工具 payload 硬性上限固定 token 预算——context window 内的静默截断是幻觉最常见原因之一：
```python
MAX_TOOL_TOKENS = 3000  # 保守上限

def truncate_tool_output(raw_output, max_tokens=MAX_TOOL_TOKENS):
    tokens = encode(raw_output)
    if len(tokens) > max_tokens:
        return decode(tokens[:max_tokens]) + "\n[TRUNCATED]"
    return raw_output
```

## Philosophy
> 你可以忍受不完美的提示词。你可以忍受天真的工具集成。但是你无法忍受一个失败时破坏自己状态或静默吞噬错误的 Agent。

## Source
- [[Your-AI-Isn-t-Stupid---It-Just-Needs-a-Better-Harness--Lychee-Technology-Engineering-Blog]]

## See Also
- [[Harness-Engineering]] — 完整学科框架
- [[7-Layer-Harness-Stack]] — 完整 7 层实现