AI 原生开放教材

外观

Better Harness（QoderAI）工程手册·英文原文 ↗原件 ↓仓库 ↗

References ​

Human-readable guidance lives here when it does not need fixtures, schemas, or runtime binding.

当某项指引不需要 fixture、schema 或运行时绑定时，面向人阅读的指引就放在这里。

Switchboard ​

  • session-evidence/: bounded session collection, Task Episode diagnostics, reader-facing session insights, and usage-efficiency evidence.
  • project-harness/: static project evidence, core-change inspection, observability, design contracts, acceptance controls, and recovery safeguards.
  • agent-customize/: agent instructions, Skills, MCP, Memory, Hooks, Custom Agents, Plugins, and platform-specific asset guidance. Start with agent-customize/routing.md.
  • bootstrap/: the 0 -> 1 move for a project with no coding-agent harness yet: the specification structure contract and stack-specific spec examples. Load this when a requirement is not yet complete enough to implement against.
  • loop-engineering/: repeated-work and schedule-ready owner selection. Load this when a task may become a Skill, automation, hook, command, script, custom agent, MCP-backed loop, or rule.
  • tool-runtimes/: support contracts for runtime discovery and execution boundaries. These are not practice domains.

本目录下的领域划分：session-evidence/ 负责有界的会话采集、Task Episode 诊断、面向读者的会话洞察与使用效率证据；project-harness/ 负责静态项目证据、核心改动检查、可观测性、设计契约、验收控制与恢复保障；agent-customize/ 负责 agent 指令、Skills、MCP、Memory、Hooks、Custom Agents、Plugins 以及各平台专属资产的指引，请从 agent-customize/routing.md 入手；bootstrap/ 负责尚无 coding-agent harness 的项目从 0 到 1 的推进——specification 结构契约与各技术栈的 spec 示例，当需求还不够完整、无法据此实现时加载它；loop-engineering/ 负责重复性工作与可调度工作的所有者选择，当一个任务可能演化为 Skill、自动化、hook、命令、脚本、custom agent、由 MCP 支撑的循环或规则时加载它；tool-runtimes/ 提供运行时发现与执行边界的支持契约，它们不属于实践领域。

Detector and signal guidance stays with its owning model, executable capability, or skill-local contract. Promote shared prose only when two visible workflow consumers need the same owner; do not recreate a generic detector bucket.

检测器与信号相关的指引留在其所属的模型、可执行能力或 skill 本地契约中。只有当两个可见的工作流消费者确实需要同一个所有者时，才把共享散文提升出来；不要重建一个通用的检测器桶。

Historical Paths ​

  • references/coding-agent-practices/ was split by evidence owner. Use session-evidence/, project-harness/, or agent-customize/.
  • references/reliable-delivery/, references/ai-friendly/, references/coding-agent-observability/, and references/ai-friendly-engineering/ were folded into project-harness/.
  • references/coding-agent-practises/ was the historical misspelled path. Do not recreate it; route the document to its current evidence owner.
  • references/harness-practises/ was the historical misspelled harness path. Do not recreate it; use references/project-harness/.
  • references/harness-engineering/ was the former broad project domain. Do not recreate it; use references/project-harness/.

历史路径的归并说明：references/coding-agent-practices/ 已按证据所有者拆分，请改用 session-evidence/、project-harness/ 或 agent-customize/；references/reliable-delivery/、references/ai-friendly/、references/coding-agent-observability/ 和 references/ai-friendly-engineering/ 已并入 project-harness/；references/coding-agent-practises/ 是历史上拼错的路径，不要再创建它，请把文档路由到它当前的证据所有者；references/harness-practises/ 是历史上拼错的 harness 路径，不要再创建它，请使用 references/project-harness/；references/harness-engineering/ 是原先宽泛的项目域，不要再创建它，请使用 references/project-harness/。