← 返回 Skills 市场
maochen1980

Assess Me

作者 MaoChen1980 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
31
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install assess-me
功能描述
Run this when debugging goes in circles, results are confusing, multiple hypotheses exist, or you need a sanity check. Audits your cognition state: goal, pro...
使用说明 (SKILL.md)

Assess Me — Self-Cognition Audit

Action

Write your current thinking state to a temp file, then read it back.

  1. Answer 6 questions and write to a temp file
  2. Read it back
  3. Critically analyze what you wrote
  4. Continue based on the findings

Questions

  1. Goal — What is the task? What does "done" look like?
  2. Progress — What's done? What's pending?
  3. Gaps — What information do you need but don't have?
  4. Assumptions — What unverified beliefs are driving your approach?
  5. Blocker — What specifically is blocking you? (exact obstacle, not symptom)
  6. Recovery — If stuck, what should you do differently?

Instructions

Write tool → /tmp/assess-me.md
Content:
# Assess Me

**Goal:** \x3C1-2 sentences>
**Progress:** \x3C1-2 sentences>
**Gaps:** \x3C1-2 sentences>
**Assumptions:** \x3C1-2 sentences>
**Blocker:** \x3C1-2 sentences>
**Recovery:** \x3C1-2 sentences>

After writing, use Read tool to read /tmp/assess-me.md back. Then review critically:

  • Is "progress" just restating effort without results?
  • Are blockers specific or vague?
  • Is "Recovery" a concrete action, or just "keep trying"?
  • Are you assuming something without evidence?

Pitfalls

  • Assumptions section must NOT be empty — there are always assumptions
  • Recovery ≠ "keep trying" — name a specific next action
  • Vague blocker = you haven't found it yet
  • Read the file back — writing without re-reading skips the detachment effect

Output Convention

When done, summarize findings inline:

[assess]
Goal: ...
Blocker: ...
Next: ...
[/assess]

Source: https://github.com/MaoChen1980/think-skills Self-optimization: After using this skill, improve it based on what you learned.

安全使用建议
Use this as a lightweight debugging checklist. Before installing or invoking it, understand that it asks the agent to write a short self-audit to /tmp/assess-me.md; do not include secrets, credentials, private user content, system prompts, or detailed internal reasoning in that file, and delete it after use if the workspace is shared or sensitive.
能力评估
Purpose & Capability
The stated purpose is to help an agent pause and audit goal, progress, assumptions, blocker, and recovery path; the requested behavior matches that purpose.
Instruction Scope
The activation criteria are broad, but that is coherent for a general debugging sanity-check skill; the template asks for short structured summaries rather than secrets or unrelated data.
Install Mechanism
The package contains only a markdown SKILL.md file, with no executable scripts, install hooks, API keys, or network behavior.
Credentials
Writing to /tmp/assess-me.md is disclosed and purpose-aligned, but users should avoid including credentials, private user data, system prompts, or sensitive reasoning details.
Persistence & Privilege
The temp-file readback step creates limited local persistence; there is no evidence of hidden persistence, privilege escalation, exfiltration, or destructive behavior.
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 assess-me
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /assess-me 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
v1.0.0
- Initial release of the assess-me skill for auditing your current thinking process during stuck or confusing problem-solving. - Guides users to systematically answer 6 key questions (Goal, Progress, Gaps, Assumptions, Blocker, Recovery) and record responses to a temporary file. - Adds a review step: read back your own answers to promote critical detachment and identify mistakes or weak points. - Enforces best practices: always list assumptions, avoid vague blockers, and make recovery plans concrete. - Output includes a standardized summary block for quick reference.
元数据
Slug assess-me
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Assess Me 是什么?

Run this when debugging goes in circles, results are confusing, multiple hypotheses exist, or you need a sanity check. Audits your cognition state: goal, pro... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 31 次。

如何安装 Assess Me?

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

Assess Me 是免费的吗?

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

Assess Me 支持哪些平台?

Assess Me 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 Assess Me?

由 MaoChen1980(@maochen1980)开发并维护,当前版本 v1.0.0。

💬 留言讨论