skills

将所有机器上的 Pi 和 OMP 会话整理成主题聚合的结构化日报，可选写入 Obsidian 工作日志。

往 pi 的 models.json 新增/适配 provider 模型，含取参、思考三层合成、capture.ts 实测验证与固定。

按需求从 npm pi-package 生态检索符合的 Pi Agent 扩展/技能/主题/prompt 包，采集 npm 元数据与 GitHub 信号，质量评估后整理成表格。与 pi-trending 互补（后者做热度发现）。

Free up disk space on Arch Linux WSL through layered cleanup: cache purge, orphan removal, package audit, and large file scanning. Use whenever the user mentions disk space, cleanup, storage, package review, or wants to shrink their WSL footprint — even if they just say "空间不够", "磁盘满了", "WSL 太大了", or "clean up my system". Prefer this skill over ad-hoc cleanup commands on Arch WSL.

Search and rewrite code by AST structure across 25 languages with ast-grep (sg). Use when the task is structural - find function calls shaped like X, rewrite console.log to logger.info, find all `as any`, migrate require() to import, find empty catch blocks, missing await, codemod a class hierarchy, scan project against YAML rules. Preferred over grep when the user mentions code SHAPE or needs deterministic refactoring across many files. Triggers: 'ast-grep', 'sg', 'ast grep', 'astgrep', 'find all functions that', 'find every call to', 'codemod', 'structural search', 'pattern match code', 'replace pattern across files', 'rename function everywhere', 'find usages of pattern', 'apply this transformation to all', 'migrate X to Y'.

Audit installed Hermes Agent skills for usage frequency and generates an interactive XLSX with Chinese descriptions, dropdown decision columns, and color-coded recommendations. Supports apply mode: reads decisions from XLSX and executes cleanup (delete/disable/enable) after user confirmation, with automatic backup. Smart filtering: already-disabled skills are not re-suggested for cleanup. Uses hermes internal API for authoritative skill classification (builtin/hub/local/external). Use when the user asks about Hermes skill usage statistics, wants to clean up unused Hermes skills, needs to disable low-usage skills, or mentions "技能审计", "清理技能", "不用的技能", "哪些技能可以删", "技能热度", "技能使用频率", "audit hermes skills". This skill is specific to Hermes Agent — do not use for other agents. Make sure to load this skill whenever Hermes skill management, cleanup, or usage analysis is mentioned.

Automate chezmoi dotfiles sync — fetch remote, pull changes, detect chezmoi-level changes (source vs home), smart re-add, commit, and push. Use whenever the user wants to sync dotfiles, says "同步", "chezmoi sync", "提交 dotfiles", "备份配置", wants to record home changes into the dotfiles repo, or is about to run chezmoi git commands manually.

Guide to managing and optimizing skills in the CNife/skills repository — quality gates, creation/publishing workflow, and the sub-agent verification loop for automated skill auditing and repair. Use when working with this repository, adding or modifying skills, running bulk operations across skills, or when the user mentions skill quality, verification, or repo maintenance.

MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). Use when the user mentions commit, rebase, squash, git blame, bisect, who wrote, when was X added, find the commit that, git history, or any git-related operations.

Manage OpenCode's permission rules in opencode.jsonc — add, remove, or list auto-approval rules for Bash commands and tool invocations so the agent stops asking for confirmation on every single command. Use whenever the user wants to auto-approve, deny, or require confirmation for a shell command, even if they don't mention "permission" or "opencode.jsonc" directly. Triggers on "允许 kubectl get *", "拒绝 rm -rf", "auto-approve npm run build", "总是执行 git status", "add permission rule", "list my permissions", "查看权限", "添加权限", "移除权限", "把 X 加到允许列表", "skip confirmation for", and similar — even if the user doesn't explicitly mention OpenCode's config.

AGENTS.md 编写与优化指南，遵循渐进式披露原则。当用户创建、修改或重构 AGENTS.md，讨论 AI agent 指令结构、规则放置位置，或提到「渐进式披露」「模块化」「AGENTS.md 最佳实践」时，务必加载此 skill。即使用户只是说「帮我写个 AGENTS.md」「优化一下这个配置文件」「拆分一下规则」，也应该使用此 skill。

Audit pi agent skills usage frequency, identify unused skills, and clean up safely. Queries pi session logs for skill invocation records (skill tags + read SKILL.md calls), cross-references against globally installed skills, categorizes by usage into high/medium/low/unused tiers, and drives cleanup via bunx skills remove. Use whenever the user mentions skill management, cleanup, or usage analysis — including "技能使用频率", "哪些技能用的多", "整理技能", "删掉不用的技能", "技能审计", "skill audit", "看看技能", "用哪些技能", "used skills", "unused skills", "清理", "clean up skills". Also use when the user has many installed skills and you suspect some are unused, or when reviewing what skills are worth keeping. This is the go-to skill for any pi skill management task. Not for Hermes agent skills — use audit-hermes-agent-skills instead.

根据 OpenAI 提示词工程最佳实践，审查并优化已有 LLM 提示词。用户说“优化提示词”“改一下 prompt”“做提示词工程”“让模型输出更好”，或贴出一段提示词请求改进时，务必使用此技能。专注诊断和改写，不负责从零创建模板、生成 / 命令、写入文件或选择保存路径。

Automate adding permission rules to Qwen Code's settings.json. Use whenever the user wants to allow, deny, or require confirmation for a command in Qwen Code. Trigger on phrases like "把 X 加到允许列表"、"添加到允许"、"允许 Bash(X)"、"deny 某命令"、"permission rule"、"auto-approve this command"、"skip confirmation for"、"always allow" — even if the user doesn't explicitly mention Qwen Code's settings.json. Also use when working with Qwen Code permission configuration or troubleshooting auto-approval rules.

在会话问题解决得差不多时提醒我整理知识到 Obsidian，根据内容规模选择日记或独立文章

Evaluate, compare, recommend, discover, and install AI agent skills. Use whenever the user wants to find, assess, choose, or install skills — even if they don't explicitly say "evaluate". Triggers: "评估/找/对比/安装/推荐 skill", "skill 安全", "哪个 skill 好用", "找个技能", "给 agent 装技能", "safe skill", "evaluate/compare/install skill", or any request involving skill discovery, security assessment, installation, or cross-agent compatibility.

从多个数据源收集用户活动轨迹 → 分析并区分工作与个人内容 → 排版为 PDF → 投递到邮箱。触发词：「发日报」「生成日报」「发我的今天」「今日汇总」「日记 PDF」「每天总结」「保存日志」「生成日报发邮件」。当你发现用户需要做每日回顾、日志汇总、生成日报 PDF 并发送时触发。

周期性巡检 Nowledge Mem 知识库，分类处理后台积累的待办事件。

发现 npm 上热门和新锐的 Pi Agent 包（扩展/技能/主题/prompt 模板），计算趋势分并汇总展示。

将会话内容按主题聚合写入 Obsidian 工作日志或个人日记，并管理待办。当用户说「记录一下」「写日记」「更新待办」等要求记录时，或在会话末尾出现结构化结论/部署完成/调研成果时加载此技能，主动询问用户是否记录。

Query AI model specs (pricing, context limits, capabilities, modalities) and provider API endpoints from models.dev. Use when the user asks about model pricing/specs, needs a provider's API/base-URL/env config, or wants to list/filter/compare models across providers. 触发例：「gpt-5 的定价」 「查 XXX 支持哪些模型」「配置 Trae 用 deepseek」。

搜索路由器：基于 opencli，按话题和场景把查询路由到合适的搜索源。当用户需要从网络获取信息时（无需用户提及 opencli）加载此技能。