Interview Bank

Agent Skills in 7 Days: Turn Experience Into Reusable Capability

D5 Hand-Building a Skill Runtime: Scanning, Frontmatter Parsing, Injecting the System Prompt, Reading the Body on Demand

• Once a skill body is in context, how do you keep it effective across a long session? And would you activate skills by file read or by a dedicated tool?skill 的正文进了上下文之后，长会话里怎么保证它不失效？激活方式上文件读取和专用工具你会选哪个？

  Common in ChinaCommon overseasDeep dive#agent-skills#runtime#long-session

  How to reason about it · think before answering

  1. This is about the gap between a working demo and something you can ship. The first half is long-session failure modes, the second is the activation mechanism trade-off.
  2. Two long-session problems. Duplicate activation: the model forgets it already read the skill and selects it again, so the same instructions appear twice, wasting tokens and creating conflicts where the wording differs. Fix it with a set of already-activated names.
  3. The worse problem is compaction. Summarizing early messages can drop the skill body, and nothing errors: the model quietly reverts to its behavior without the skill. Users report that it stopped following the convention later in the conversation, and it is the hardest failure here to diagnose.
  4. The fix is to mark the activated message as protected so compaction preserves it, or to re-inject it afterward. The marker is trivial; remembering to set it is not.
  5. For the second half give criteria, not a preference. File-read activation adds no new mechanism, so any agent that can read files supports skills immediately, which is why the format spread across dozens of clients. The cost is no clean hook for dedup or protection, and the model can read the wrong path.
  6. A dedicated tool turns activation into an observable, interceptable call where you can dedupe, check permissions, and return the skill directory and resource list together. The cost is another tool definition and host cooperation. The criterion is whether you control the host.
  7. Expected follow-up: should resource files be read during activation? No, list filenames only. The value of three stages is that the third usually never happens.

  分析过程 · 先想清楚再作答

  1. 这题考的是「演示能跑」和「上线能用」之间那段距离。前半是长会话的失效模式，后半是激活机制的取舍。
  2. 长会话有两个问题。第一个是重复激活：模型忘了自己读过，第二次又选中同一个 skill，同一段指令出现两遍既浪费又容易在措辞出入时互相干扰。修法是维护一个已激活集合，命中就直接返回。
  3. 第二个问题更要命——**被压缩掉**。压缩会把早期消息换成摘要，skill 正文落在那个区间里**不会报任何错**，模型只是悄悄退回没有这个 skill 的行为。用户看到的现象是「聊到后面它又不按规范写了」，这是这套机制里最难查的一类问题。
  4. 解法是给激活出来的那条消息打一个受保护标记，压缩时整段保留，或者在压缩后重新注入一次。标记本身很简单，难的是记得给它。
  5. 后半的取舍要给判据而不是偏好。文件读取式零新增机制，任何有读文件能力的 Agent 都能立刻支持，这正是这个格式能在几十家客户端铺开的原因；代价是没有明确钩子做去重和保护，模型还可能读错路径。
  6. 专用工具式把激活变成一次可观测可拦截的调用，能在这一步做去重、权限检查、连技能目录与资源清单一起返回；代价是多一个工具定义，且要求宿主愿意开这条通路。**判据是你控不控得住宿主**：自己写 Agent 用工具式，做通用实现用文件读取式。
  7. 可预期的追问是「资源文件要不要在激活时一起读进来」。不要，只列文件名。三阶段的全部价值就在于第三阶段大多数时候不会发生。

  Key points

  • Dedupe with a set of activated skills or the same instructions appear twice and conflict.
  • Losing a skill body to compaction raises no error; the model silently reverts, which is the hardest failure to spot.
  • Mark the activated message as compaction-protected, or re-inject after compaction.
  • File-read activation adds no mechanism and has the best compatibility but offers no hook for dedup or protection.
  • A dedicated tool is observable and interceptable; choose by whether you control the host, and in both cases list resource filenames without reading them.

  答题要点

  • 重复激活要靠已激活集合去重，否则同一段指令会出现两遍并互相干扰。
  • 压缩掉 skill 正文不会报错，模型只会悄悄退回原行为，是最难查的失效。
  • 激活出来的消息要打受保护标记，压缩时保留或事后重新注入。
  • 文件读取式零新增机制、兼容性最好，但没有去重与保护的钩子。
  • 专用工具式可观测可拦截，判据是你控不控得住宿主；两者都只列资源文件名，不预读内容。