← 返回 Skills 市场
lakendocean

Idea Trace

作者 ken · GitHub ↗ · v0.1.1 · MIT-0
cross-platform ✓ 安全检测通过
123
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install idea-trace
功能描述
Don't just save ideas - trace how they mature. Use when a user wants to follow how an important idea was shaped by later conversations, work, projects, and l...
使用说明 (SKILL.md)

Idea Trace

Don't just save ideas - trace how they mature.

Good ideas rarely arrive complete. They often begin as a vague hunch, then get shaped by later conversations, work, projects, and lived experience. Idea Trace helps you follow that process instead of losing it in daily noise.

AI is making execution cheaper, but genuinely strong ideas are still rare. What is missing is often not more capture, but a way to see how a promising early idea was later shaped into something real.

This is not an idea inbox. It is for following how important ideas evolve.

Before substantive work, read references/taxonomy.md.

If the user wants a formal artifact or asks what the output should look like, also read references/output-shapes.md.

Use This Skill When

  • the user wants to trace how an idea, stance, method, or judgment became mature
  • the user wants to connect scattered later events back to an earlier important idea
  • the user wants help deciding whether something is a trackable core proposition or just a passing idea
  • the user wants to know what each later contribution changed, not just what happened
  • the user wants to surface an ongoing long-term line, not only explain the past

Do not use this skill for:

  • one-off task ideas or fleeting inspiration
  • generic note cleanup or simple archiving
  • plain timeline summaries with no shaping logic
  • graph linking that does not explain contribution type or effect
  • platform automation or workflow design unless the real task is still proposition tracing

What Good Looks Like

The skill succeeds only when the output makes these questions easy to answer:

  • what is the core proposition behind the idea
  • why is it worth tracking
  • which later events or judgments actually shaped it
  • what each contribution changed
  • what the current mature formulation is
  • what still remains open, unstable, or active

Workflow

  1. Qualify the object.
    • Test whether it is a true core proposition using references/taxonomy.md.
    • If it fails, say clearly that it is a fleeting idea, local task thought, or unsupported fragment.
  2. Define the proposition spine.
    • Capture the earliest seed form, current form, scope, and why it matters.
  3. Gather candidate contributions.
    • Look across life, work, projects, and dialogue.
    • Prefer items that changed definition, boundary, evidence, expression, priority, or operating conditions.
  4. Classify shaping roles.
    • Label each contribution using the contribution-type taxonomy.
    • Reject items that are only adjacent or repetitive.
  5. Build the evolution chain.
    • Organize by turning points, not raw chronology.
    • State what changed at each step and why it mattered.
  6. Produce the right artifact.
    • Use references/output-shapes.md when the user needs a map, narrative, dossier, or active-line snapshot.
  7. Mark confidence honestly.
    • Separate confirmed links from inference.
    • Mark weak, partial, or speculative connections explicitly.

Core Rules

  • Speak publicly in terms of "important ideas," but analyze them as core propositions.
  • Track core propositions, not every idea the user has ever had.
  • Explain shaping effect, not just relatedness.
  • Prefer turning points over exhaustive timelines.
  • Preserve both retrospective explanation and ongoing-line recognition when evidence supports both.
  • If a link is weak, say it is weak.
  • If the conversation starts drifting into implementation too early, stabilize the conceptual model first.

Trigger Examples

Use this skill for asks like:

  • "Help me show how this idea became a mature judgment."
  • "Which later projects and conversations were actually shaping this thesis?"
  • "Is this worth tracking as a core proposition or is it just a passing thought?"
  • "Map how this concept was gradually corrected, expanded, and reframed."
  • "Show me the long-term line I have been pushing without fully noticing it."
安全使用建议
This skill appears internally consistent and does not request credentials or install code. Before using it, be mindful of the data you provide: Idea-tracing requires feeding the agent your notes, conversations, or project materials, which may contain sensitive information. Do not paste secrets (passwords, private keys, or private PII) into the chat. If you plan to connect this skill to a memory store or external document system, review that integration's permissions and audit logs first. Finally, if you want absolute assurance, check that the deployed agent runtime will not automatically share your data with third-party services before invoking the skill.
能力评估
Purpose & Capability
Name/description (tracing idea evolution) align with the SKILL.md and the included reference files; nothing in the package requests unrelated capabilities (no external APIs, binaries, or secrets).
Instruction Scope
Runtime instructions are narrowly scoped to qualifying a proposition, gathering candidate contributions, classifying them, and producing outputs using the included references. It explicitly tells the agent to read local files (references/taxonomy.md and references/output-shapes.md), which are present. Minor note: agents/openai.yaml's default_prompt contains a $idea-trace placeholder variable (likely harmless), but otherwise the instructions do not instruct the agent to read unrelated files, env vars, or external endpoints.
Install Mechanism
No install spec and no code files to write or execute; instruction-only skill is lowest-risk in install mechanism terms.
Credentials
The skill declares no required environment variables, credentials, or config paths — proportional to its stated purpose.
Persistence & Privilege
always is false, autonomous model invocation is allowed (platform default) and appropriate for this skill. The package does not request permanent presence or modify other skills or system settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install idea-trace
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /idea-trace 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v0.1.1
Sharpened the public-facing copy for stronger resonance and clearer first-screen positioning.
v0.1.0
Initial release of Idea Trace for tracking how rare high-value ideas evolve into mature views.
元数据
Slug idea-trace
版本 0.1.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

Idea Trace 是什么?

Don't just save ideas - trace how they mature. Use when a user wants to follow how an important idea was shaped by later conversations, work, projects, and l... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 123 次。

如何安装 Idea Trace?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install idea-trace」即可一键安装,无需额外配置。

Idea Trace 是免费的吗?

是的,Idea Trace 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Idea Trace 支持哪些平台?

Idea Trace 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Idea Trace?

由 ken(@lakendocean)开发并维护,当前版本 v0.1.1。

💬 留言讨论