By simplecyon
A structured thinking skillset — clarifier, muse, thinker, janus, conductor
Use when the user's request is fuzzy or hedged ('something like', 'I think', 'not sure', 有个想法, 大概是, 说不清, 还没想清楚); when a goal is stated but scope, audience, or done-condition is undefined; when the user asks where to start (不知道从哪开始, 帮我理一理, 理一下思路); when they've been going back and forth and need help landing; or when what they *want* and what they *asked for* seem misaligned. Do NOT wait for the user to say they're confused — if the request itself is ambiguous, this skill should activate. /clarifier explicitly triggers this.
Use when thinking has gone on for multiple rounds and needs converging: the user says 收一下, 怎么取舍, 接下来怎么办, 有点乱, 想太多了, 聊了这么多, 'so where does that leave us', 'what should I do next'; the conversation has produced multiple angles or analyses but no clear direction; the user seems overwhelmed by too many good ideas; or thinking is going in circles. /conductor explicitly triggers this.
Use when the user presents a conclusion, plan, decision, or recommendation that needs stress-testing, or says 我认为, 结论是, 应该, 最好的方案, 只需要, 已经证明, 肯定, 显然, 质疑一下, 有没有问题, 帮我挑挑毛病. /janus explicitly triggers this.
Use when the conversation is going in circles or converging too quickly on one direction; when the user is stuck or wants fresh angles (没思路, 卡住了, 换个角度, 有没有别的可能, 发散一下); when a plan or idea feels off or flat (太平了, 不够有意思); when the user is exploring anything creative, strategic, or conceptual (product thinking, content, positioning, naming, pitches) and hasn't asked for deep analysis; or when the user seems to want validation but would benefit more from a provocation. /muse explicitly triggers this.
Use when the user wants one angle thought through deeply rather than more ideas: they share a substantial block of thinking or notes and want insight, not a summary (想透, 深入分析); they face a decision with real tradeoffs (纠结, 利弊, 帮我想想这个决定); they want assumptions questioned or blind spots surfaced (盲区); they want to understand *why* something is the way it is (为什么会这样); or they name a specific framework (first principles, second-order, inversion, 第一性原理, 逆向思考, 主要矛盾). /thinker explicitly triggers this.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
一套结构化的思维技能包——不是给你更多答案,而是帮你在对的时刻用对的方式思考。
MindForge 是一个思维方法论技能包,由 5 个互补的角色 + 1 个入口 skill 组成:
| Skill | 角色 | 核心动作 |
|---|---|---|
clarifier | 澄清者 | 把模糊聊成清晰 |
muse | 灵感触发器 | 给新角度和问题 |
thinker | 深度分析 | 把一个角度想透 |
janus | 批判审查 | 压力测试结论 |
conductor | 指挥 | 收拢思维、做取舍 |
using-mindforge | 入口 | 调度以上角色 |
模糊 → 清晰 → 发散 → 深入 → 质疑 → 收拢
│ │ │ │ │ │
│ │ │ │ │ └─→ /conductor
│ │ │ │ └──────────→ /janus
│ │ │ └──────────────────→ /thinker
│ │ └──────────────────────────→ /muse
│ └──────────────────────────────────→ (可以开工了)
└──────────────────────────────────────────→ /clarifier
不是每段对话都需要走完全部流程。知道什么时候用什么,比什么都用更重要。
/plugin marketplace add simplecyon/agentkit
/plugin install mindforge@agentkit
安装后,SessionStart hook 会在每次新会话开始时自动注入 using-mindforge 入口 skill——agent 不需要你手动触发就知道什么时候该用哪个角色。
npx skills add simplecyon/mindforge-skills -g -y
git clone https://github.com/simplecyon/mindforge-skills.git
# 将 skills/ 目录下的 skill 复制到你的 agent 的 skills 目录
安装后,在对话中使用斜杠命令触发:
/clarifier — 需求模糊时/muse — 需要新角度时/thinker — 要深入分析时/janus — 有了结论要质疑时/conductor — 想太多了要收拢时每个 skill 有它的时刻。先澄清,再发散,再深入,再质疑,最后收拢。顺序可以跳,但不能乱。
最好的思考是被问出来的。这些 skill 不会给你答案,会给你更好的问题。
每个 skill 结束时,如果判断该换角色了,会自然建议下一个。不打断节奏,不强推。
当思维进行了多轮,conductor 负责把散落的思维收拢成有用的东西——定位、蒸馏、取舍、方向。
MindForge 专注于思维层面——澄清、创意、分析、批判、决策。
Superpowers 专注于执行层面——计划、TDD、debug、code review。
两者互补,可以同时安装。
MIT
npx claudepluginhub simplecyon/agentkit --plugin mindforgePermanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
Intelligent prompt optimization: injects the right context at the right moment so Claude lands a better first output. Clarifies vague prompts with research-based questions, plus targeted nudges for approach selection, plan readability, workflow routing, background execution, subagent routing, output readability, user-decision questions, and plan-mode assessment
Semantic search for Claude Code conversations. Remember past discussions, decisions, and patterns.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows