← 返回 Skills 市场
598
总下载
2
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install competing
功能描述
Improve systematically by analyzing losses, studying winners, and tracking progress against rivals in any competitive domain.
使用说明 (SKILL.md)
Core Framework
Competition is a learning accelerator. Every loss contains the lesson that wins don't.
- Analyze the Loss — Don't just lose, understand WHY
- Study the Winner — What did they do that you didn't?
- Track the Delta — Measure the gap, watch it shrink
- Iterate — Apply lessons, compete again, repeat
The Post-Loss Protocol
After any competitive loss, extract value:
| Question | Purpose |
|---|---|
| At what moment did the outcome shift? | Find the decision point |
| What did they do that I didn't? | Identify the winning move |
| What would I do differently? | Formulate the lesson |
| Is this a pattern? | Check history for repeats |
Don't rationalize. Don't blame externals. Extract the actionable insight.
Tracking (What to Measure)
Create a tracking folder in the user's workspace:
~/competing/
├── domains/ # Per-domain tracking
├── rivals.md # Opponent profiles
├── log.md # Win/loss log with lessons
└── progress.md # Metrics over time
For each domain, track:
- Win/loss record with dates
- Specific losses analyzed (who, why, lesson)
- Patterns identified (recurring weaknesses)
- Progress metrics (are lessons translating to wins?)
Rival Intelligence
Know your competition:
- Profile rivals — Their strengths, weaknesses, tendencies
- Monitor changes — When they improve or change strategy
- Find their edge — What specifically makes them beat you?
- Study up — Find examples of them losing, analyze what worked
Quick Reference
| Situation | Action |
|---|---|
| Just lost | Run post-loss protocol, add to log |
| Pattern emerging | Document it, create drill/fix |
| Preparing for known rival | Review their profile, past matches |
| Plateau in progress | Analyze recent losses for new patterns |
| Won against usual winner | Document what changed, replicate |
Load Reference
| Need | File |
|---|---|
| Domain-specific strategies | domains.md |
| Deep loss analysis framework | analysis.md |
| Progress tracking templates | tracking.md |
| Feedback loop mechanics | feedback.md |
安全使用建议
This is an instruction-only improvement framework and appears coherent and low-risk. Before installing or enabling: (1) acknowledge the skill recommends creating ~/competing/ and markdown logs — avoid putting sensitive data (passwords, private keys, or confidential competitor data) into those files; (2) confirm your agent/runtime has permission to write to your workspace if you want the files created; (3) note there is no network activity or external installs expected, so risks are limited to local file creation and any data you choose to record. If you want stricter control, keep logs in a private location or avoid storing sensitive details.
功能分析
Type: OpenClaw Skill
Name: competing
Version: 1.0.0
The skill bundle is benign. Its purpose is to help users track and improve competitive performance, and all instructions and content align with this goal. The only file system interaction instructed is the creation of a user-local tracking folder (`~/competing/`) and associated markdown files, which is a legitimate and expected function for such a skill. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection attempts designed to subvert the agent for harmful purposes.
能力评估
Purpose & Capability
Name/description (improve by analyzing losses, studying winners, tracking rivals) matches the contents: analysis templates, tracking formats, domain guidance and rivals intelligence. No unrelated credentials, binaries, or external services are requested.
Instruction Scope
Runtime instructions are limited to analysis prompts, templates, and a recommendation to create a ~/competing/ workspace and markdown files. There are no instructions to read system secrets, call external endpoints, or access unrelated system paths.
Install Mechanism
No install spec and no code files — instruction-only. This minimizes on-disk execution risk; nothing is downloaded or installed.
Credentials
The skill declares no environment variables, credentials, or config paths. The guidance to store logs and rival profiles locally is proportional to the claimed tracking purpose.
Persistence & Privilege
always:false and normal model invocation behavior. The skill does not request persistent system privileges or modifications to other skills; suggested local files are self-contained in the user's workspace.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install competing - 安装完成后，直接呼叫该 Skill 的名称或使用
/competing触发 - 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
元数据
常见问题
Competing 是什么？
Improve systematically by analyzing losses, studying winners, and tracking progress against rivals in any competitive domain. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 598 次。
如何安装 Competing？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install competing」即可一键安装，无需额外配置。
Competing 是免费的吗？
是的，Competing 完全免费（开源免费），可自由下载、安装和使用。
Competing 支持哪些平台？
Competing 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 Competing？
由 Iván（@ivangdavila）开发并维护，当前版本 v1.0.0。
推荐 Skills