typo update

This commit is contained in:
rookit
2026-04-07 10:59:29 +08:00
parent 7868f554ef
commit deb9da2623
8 changed files with 12 additions and 3 deletions
@@ -31,6 +31,8 @@ An alert is secondary data in ASP. Each alert belongs to a case, and each alert
- If the user is working on a specific alert, prefer `list_alerts(alert_id=<id>, limit=1)` because the current MCP surface does not expose a separate `get_alert` tool.
- If the user wants to save structured analysis back onto the alert, use the `asp-enrichment-en` skill.
Note: alerts only support `severity_ai`, `confidence_ai`, and `comment_ai` updates; verdict and summary fields belong to case updates.
## Additional Information
- `rowid` is the UUID for each alert record and is used for data association.
@@ -15,7 +15,7 @@ metadata:
# ASP Artifact
Use this skill when the user needs to investigate artifacts on ASP.
An artifact is the third-level data in ASP. Each artifact is attached under one alert and is the smallest data unit that still needs investigation.
Artifacts are created automatically by system processes. You can list and analyze existing artifacts, and save analysis back through enrichment.
## When to Use
@@ -15,7 +15,7 @@ metadata:
# ASP Artifact
当用户要围绕 artifact 进行调查分析时,使用这个 skill。
artifact 是 ASP 中的三级数据,每个 artifact 都挂载在一个 alert 下,是最小的需要调查的数据单元
artifact 由系统自动创建,用户只能查询和分析已有 artifact,并通过 enrichment 保存分析结果
## 适用场景
@@ -44,6 +44,8 @@ metadata:
3. 如果用户已经有现成的 enrichment rowid,跳过创建,直接附加。
4. 如果用户还处于对象探索阶段而不是保存结果,先使用对应对象 skill。
当你已经有明确的分析结论,例如 verdict、TTP 集合、风险评级或缓解建议,并且这些内容需要保存在目标对象上时,就切换到这个 skill。
## SOP
### 创建并附加新的 Enrichment
@@ -48,6 +48,8 @@ ASP internal Knowledge is stored as individual database records. The main fields
- When the user says to "add to the knowledge base" or "make it retrievable", interpret that as setting `action` to `Store`, not editing `using` directly.
- When the user says to "remove from the knowledge base" or "stop using it for retrieval", interpret that as setting `action` to `Remove`.
Note: `action=Store` and `action=Remove` are handled asynchronously by the backend, so the retrievability state may lag behind the update.
## Decision Flow
1. If the user wants to search vectorized knowledge content for relevant conclusions, experience, or historical knowledge, use `search_knowledge`.
@@ -48,6 +48,8 @@ ASP 的内部 Knowledge 本质上是一条一条数据库记录,核心字段
- 当用户说“加入知识库”或“让它可被检索”时,通常应理解为把 `action` 设为 `Store`,而不是直接改 `using`
- 当用户说“移出知识库”或“不要再参与检索”时,通常应理解为把 `action` 设为 `Remove`
注意:`action=Store``action=Remove` 由后端异步处理,所以可检索状态可能会滞后于更新。
## 决策流程
1. 如果用户要在“已向量化的知识内容”里查找相关结论、经验或历史知识,使用 `search_knowledge`
@@ -95,6 +95,8 @@ Preferred response structure:
- Prefer workflow wording over storage wording.
- When many tickets match, show the best subset and explain the pattern briefly.
Tip: there is no direct `case_id` filter for `list_tickets`; use `list_cases(case_id=..., lazy_load=false)` to inspect attached tickets when needed.
## Failure Handling
- If no tickets match, say that directly and suggest the most useful refinement.
-1
View File
@@ -439,7 +439,6 @@ def update_knowledge(
title=title,
body=body,
action=action,
source=source,
tags=tags
)