Back to Details

turbocharge

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

MANDATORY PREPARATION

必备准备工作

Invoke {{command_prefix}}agent-workflow — it contains workflow principles, anti-patterns, and the Context Gathering Protocol. Follow the protocol before proceeding — if no workflow context exists yet, you MUST run {{command_prefix}}teach-maestro first.
Start your response with:
text
──────────── ⚡ TURBOCHARGE ─────────────
》》》 Entering turbocharge mode...
Push a workflow past conventional limits. This isn't about adding features — it's about making existing capabilities operate at a level users didn't think was possible.
EXTRA IMPORTANT: Context determines what "extraordinary" means. Understand the project's scale before deciding what to turbocharge.
调用{{command_prefix}}agent-workflow —— 它包含工作流原则、反模式以及上下文收集协议。在继续操作前请遵循该协议——如果还没有任何工作流上下文，你必须先运行{{command_prefix}}teach-maestro。
你的回复开头请使用以下内容：
text
──────────── ⚡ TURBOCHARGE ─────────────
》》》 Entering turbocharge mode...
突破工作流的常规限制。这不是要新增功能，而是让现有能力达到用户原本认为不可能实现的运行水平。
特别重要：上下文决定了「超乎寻常」的定义。在决定要优化哪些部分的性能之前，先了解项目的规模。

Propose Before Building

构建前先提出方案

  1. Think through 2-3 different directions with trade-offs
  2. Present these options to the user and wait for their selection before writing code
  3. Only proceed with the confirmed direction
  1. 梳理2-3种不同的实现方向并明确各自的取舍
  2. 将这些选项呈现给用户，等待用户选择后再编写代码
  3. 仅按照确认后的方向继续推进

For high-throughput workflows

适用于高吞吐量工作流的优化方案

  • Parallel fan-out: Split input, process N simultaneously, merge results
  • Streaming pipelines: Start processing step N+1 while step N runs
  • Progressive quality: Fast pass on everything, detailed pass on flagged items
  • Smart batching: Group similar items, outliers get individual treatment
  • 并行扇出：拆分输入，同时处理N个任务，再合并结果
  • 流式管道：在第N步运行时就启动第N+1步的处理
  • 渐进式质量控制：所有内容先快速过一遍，仅对标记项做详细处理
  • 智能分批：将相似项分组，异常项单独处理

For latency-critical workflows

适用于延迟敏感型工作流的优化方案

  • Speculative execution: Start likely next step before current finishes
  • Cached warm paths: Pre-compute responses for common patterns
  • Model cascading: Try fastest model first, escalate only when needed
  • 推测执行：在当前步骤完成前就启动大概率会执行的下一步
  • 缓存热路径：提前计算常见模式的响应结果
  • 模型级联：优先尝试最快的模型，仅在必要时升级使用更复杂的模型

For reliability-critical workflows

适用于可靠性敏感型工作流的优化方案

  • Automatic failover: Detect failures, switch to alternatives automatically
  • State checkpointing: Save state, resume from any point after crash
  • Chaos testing: Intentionally break dependencies to verify recovery
  • 自动故障转移：检测故障，自动切换到备选方案
  • 状态检查点：保存状态，崩溃后可从任意节点恢复
  • 混沌测试：故意破坏依赖项以验证恢复能力

For adaptive workflows

适用于自适应工作流的优化方案

  • Complexity routing: Route simple inputs to fast paths, complex to thorough
  • Dynamic model selection: Choose model based on task requirements
  • Feedback-driven optimization: Track what works best, adapt routing
  • 复杂度路由：将简单输入路由到快速路径，复杂输入路由到完整处理路径
  • 动态模型选择：根据任务需求选择模型
  • 反馈驱动优化：跟踪最优方案，调整路由规则

Progressive enhancement is non-negotiable

渐进式增强是硬性要求

Every turbocharge technique must degrade gracefully. The workflow without the enhancement must still work.
所有性能优化技术都必须具备优雅降级能力。没有增强功能的工作流仍需正常运行。

Verification

验证

  • Performance test: Is it measurably faster/cheaper/more reliable?
  • Degradation test: Disable enhancement — does it still work?
  • Cost test: Does improvement justify complexity?
  • Maintenance test: Can someone else maintain this in 6 months?
  • 性能测试：是否在速度/成本/可靠性上有可量化的提升？
  • 降级测试：禁用增强功能后，工作流是否仍能正常运行？
  • 成本测试：提升的收益是否匹配新增的复杂度？
  • 可维护性测试：其他人员是否能在6个月后仍能维护这套方案？

Recommended Next Step

推荐后续步骤

After turbocharging, run 
{{command_prefix}}evaluate
to verify the enhancement works and degrades gracefully.
NEVER:
  • Turbocharge before the workflow is correct (make it right, then make it fast)
  • Add complexity without measuring the improvement
  • Build self-healing without testing the healing
  • Layer multiple turbocharge techniques at once
完成性能优化后，运行
{{command_prefix}}evaluate
来验证增强功能是否正常运行，且可优雅降级。
严禁：
  • 在工作流本身正确性未验证前就做性能优化（先做对，再做快）
  • 不衡量提升效果就新增复杂度
  • 不测试恢复能力就构建自修复功能
  • 一次性叠加多种性能优化技术