文档生成
- 作者仓库星标 0
- 作者更新于 实时读取
- 作者仓库 plugins
- 领域
- 通用
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @0-xcf · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 即装即用
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- 未声明（默认跨平台）
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生，可能与作者实际意图存在差异。
需要注意： 未限定 allowed-tools，默认拥有全部工具权限。
---
name: maintain-changelog
description: | they're done, append an entry to CHANGELOG.md in the repo root. Also activate when: the diff t…
category: 通用
runtime: 无特殊运行时
---
# maintain-changelog 输出预览
## PART A: 任务判断
- 适用问题：通用任务拆解、检查和交付。
- 输入要求：目标材料、限制条件、期望输出和验收方式。
- 证据边界：围绕“When to Activate / Changelog Location / Entry Format”读取原文规则，不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断：确认你的需求是否属于通用任务拆解、检查和交付，并标出输入、限制和预期结果。
- **02** 执行计划：优先按“When to Activate / Changelog Location / Entry Format”拆成步骤，说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果：给出可复制的命令、文件改动、检查清单或内容草稿，并说明如何继续迭代。
- **04** 风险边界：结合 读取文件、写入/修改文件、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件；主要在本地完成；通常不需要额外 API Key。
- 先小样例验证，再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 先确认触发方式
原文没有稳定的斜杠命令要求。安装验证后通常全局生效，直接在对话里点名这个 Skill 并描述任务即可。
给清楚输入和边界
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是：读取文件、写入/修改文件。
小样例验证后再放大
先用一个小任务确认它会围绕“When to Activate / Changelog Location / Entry Format”工作；涉及文件或命令时，先看 diff、日志、预览或测试结果。
复核后再交付
检查最终产物是否包含明确结果、必要证据和下一步动作；如果输出泛泛而谈，就补充输入、边界和验收标准后重跑。
---
name: maintain-changelog
description: | they're done, append an entry to CHANGELOG.md in the repo root. Also activate when: the diff t…
category: 通用
source: 0-xcf/plugins
---
# maintain-changelog
## 什么时候使用
- maintain-changelog 是一个通用扩展技能，按 SKILL 适合处理通用任务拆解、检查、交付和复盘，核心价值是把输入、判断、执行、验证和交付边界固定下来，避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代的…
- 面向通用任务拆解、检查和交付，优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「When to Activate / Changelog Location / Entry Format」组织步骤，不把推断写成作者事实。
- 读取文件、写入/修改文件；主要在本地完成；通常不需要额外 API Key。
- 先跑小样例，确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown，不编造命令、平台或依赖。 证据边界与执行链路
作者原文负责流程事实；仓库文件负责来源和命令；流狐只补充适用场景、限制和质量判断。
skill "maintain-changelog" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> When to Activate / Changelog Location / Entry Format
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
}Maintain Changelog
When to Activate
End of every session that modified files. Before wrapping up or when the user signals
they're done, append an entry to CHANGELOG.md in the repo root.
Also activate when:
- The user explicitly asks to update the changelog
- Starting a new session — read the changelog first to understand recent history
Changelog Location
CHANGELOG.md in the repository root. Create it if it doesn't exist.
Entry Format
## YYYY-MM-DD — <brief title>
### What changed
- Bullet points describing each meaningful change
- Group related changes together
- Include file paths for significant additions/modifications
### Why
- One or two sentences explaining the motivation or context
### Files touched
- `path/to/new-file.yaml` (new)
- `path/to/modified-file.yaml` (modified)
- `path/to/deleted-file.yaml` (deleted)Rules
- Be concise. Each entry should be scannable in 10 seconds. No paragraphs.
- Focus on "what" and "why", not "how". Don't describe implementation details — the diff tells that story. The changelog tells the human story.
- Group by logical change, not by file. "Added shared Postgres instance" is one entry even if it touched 3 files.
- Newest entries at the top. Reverse chronological order.
- Don't log trivial changes. Typo fixes, formatting, or changelog-only updates don't need entries.
- Include context that won't be obvious later. Item IDs, IP addresses, design decisions — things that help future sessions understand the state of the world.
- Don't duplicate commit messages. The changelog is higher-level than git log. One changelog entry may span multiple commits.
Example
# Changelog
## 2026-02-09 — Add shared PostgreSQL instance
### What changed
- Created CNPG Cluster `home-1` in `postgres` namespace (single instance, 10Gi on zfs-nfs)
- Added managed roles for `kutt` and `linkding` with ESO-managed passwords (auto-rotation via CNPG)
- Created 4 Vaultwarden login items for postgres credentials (superuser, app, kutt, linkding)
- Added ArgoCD Application for postgres
### Why
Shared Postgres instance for apps migrating off SQLite. Per-app user isolation with
automated password rotation through the ESO → CNPG pipeline.
### Files touched
- `apps/postgres/postgres.yaml` (new)
- `apps/external-secrets/externalsecrets.yaml` (modified — 4 new ExternalSecrets)
- `apps/argocd/apps.yaml` (modified — postgres Application added)
### Next steps
- Create `kutt_db` and `linkding_db` databases manually after cluster is healthy
- Rotate change-me passwords in Vaultwarden
- Migrate Kutt from SQLite to PostgresStarting a New Session
When beginning work, read CHANGELOG.md to understand:
- What was done recently
- What "next steps" were left by the previous session
- The current state of the project
This provides continuity across sessions without relying on memory files alone.
Notes
- The changelog complements (not replaces) git history. Git shows what code changed; the changelog shows what decisions were made and why.
- If the repo doesn't have a CHANGELOG.md yet, create one with a header:
# Changelog - Don't commit the changelog separately — include it in the same commit as the changes it documents.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核