← 返回 Skills 市场

Agent Compete Scope

Name: Agent Compete Scope
Author: jinu4you

作者 jinu4you · GitHub ↗ · v1.0.2

cross-platform ⚠ suspicious

299

总下载

当前安装

版本数

在 OpenClaw 中安装

/install agent-compete-scope

功能描述

경쟁사 포지셔닝 분석 및 화이트스페이스 도출 에이전트

使用说明 (SKILL.md)

🔍 CompeteScope Agent

내 제품과 경쟁사들을 비교 분석하여 시장의 빈틈(Whitespace)과 차별화 전략을 제안합니다.

Features

경쟁사 프로필: 웹 검색을 통한 경쟁사 상세 프로필 자동 생성
비교 매트릭스: 기능, 가격, 타겟 등 다차원 비교표 생성
전략 도출: AI가 분석한 시장 기회 및 추천 전략 제공

Usage

ACP Job Payload:

{
  "my_product": "AI 마케팅 툴",
  "competitors": ["Competitor A", "Competitor B"]
}

安全使用建议

This skill implements competitor analysis via web search and LLM calls, which fits its description, but the package metadata omits important details. Before installing or running: 1) Do not supply API keys until you confirm which provider you'll use (default is GROQ via GROQ_API_KEY; Tavily is used for web search via TAVILY_API_KEY). 2) Expect outbound network calls to api.tavily.com, generativelanguage.googleapis.com (if Gemini/Google selected), api.groq.com, and Anthropic endpoints; review those endpoints and be comfortable with sending scraped content and competitor names there. 3) The repository contains package.json and package-lock — run npm install/build in an isolated environment (or review code) before executing. 4) The SKILL.md and registry metadata don't list required env vars; ask the publisher to update metadata to list required secrets and an install guide. 5) If you must test, run with mock mode (TAVILY_API_KEY unset) and without real secrets, and inspect network calls/logs. If you are not comfortable exposing API keys or data to third-party LLM/search services, do not install/use this skill.

功能分析

Type: OpenClaw Skill Name: agent-compete-scope Version: 1.0.2 The CompeteScope bundle is a legitimate competitor analysis tool that uses the Tavily API for web searching and various LLM providers (Groq, Anthropic, Gemini) for data synthesis. The code logic in src/analyzer.ts and src/fetcher.ts is well-structured, follows the stated purpose in SKILL.md, and contains no evidence of data exfiltration, malicious execution, or prompt injection. All external network calls are directed to official API endpoints (api.tavily.com, api.groq.com, etc.), and dependencies in package.json are standard industry libraries.

能力评估

⚠ Purpose & Capability

The skill's stated purpose (competitor positioning and whitespace analysis) is consistent with the code (web searching + LLM analysis). However the registry metadata declares no required environment variables or binaries while the code and README require API keys (TAVILY_API_KEY, GROQ_API_KEY, and optionally ANTHROPIC_API_KEY or GOOGLE_API_KEY). This mismatch is incoherent and could lead users to install/run the skill without realizing it will use external services and secrets.

⚠ Instruction Scope

SKILL.md shows only the job payload and feature description and does not mention loading .env or external APIs, yet the runtime code (fetcher + LLM adapters) reads .env and sends data to external endpoints (api.tavily.com, generativelanguage.googleapis.com, api.groq.com, Anthropic SDK). Sending scraped content and competitor names to external LLM/search APIs is within the stated purpose but the instructions omit these data flows and do not warn about secrets or outbound network calls.

ℹ Install Mechanism

There is no install spec in the registry, but the package.json/package-lock are included and declare npm dependencies (@anthropic-ai/sdk, openai, dotenv, etc.). The dependency sources are standard npm packages (no obscure download URLs), so installation risk is typical for an npm project — but the absence of an install spec in registry metadata is an inconsistency that should be clarified (the project requires npm install/build to run).

⚠ Credentials

The code expects multiple API keys (GROQ_API_KEY by default via LLM_FACTORY, plus optional ANTHROPIC_API_KEY or GOOGLE_API_KEY and TAVILY_API_KEY for web search). The registry declares no required env vars or primary credential. While these keys are logically related to the skill's function, the omission from metadata makes the credential requests non-transparent. Also multiple provider keys increase the secret surface; confirm you only provide the keys you intend this skill to use.

✓ Persistence & Privilege

The skill does not request always:true, does not modify other skills or system-wide configs, and has no install spec that writes unexpected files. It will load a local .env and may create network traffic, but it does not ask for elevated system privileges. Autonomous invocation is allowed (platform default) but not combined with 'always: true'.

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 agent-compete-scope
安装完成后，直接呼叫该 Skill 的名称或使用 /agent-compete-scope 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

v1.0.2

- Updated package.json only; no changes to features or documentation. - No functional changes in this version.

v1.0.1

- Updated package.json with version 1.0.1. - No changes to functionality or documentation content.

v1.0.0

Initial release of CompeteScope Agent. - 경쟁사 프로필 자동 생성 (웹 검색 기반) - 제품별 기능, 가격, 타겟 등 비교 매트릭스 생성 - 시장의 빈틈(Whitespace)과 차별화 전략 제안 - 사용 예시 및 ACP Job Payload 가이드 추가

元数据

Slug agent-compete-scope

版本 1.0.2

许可证 —

累计安装 1

当前安装数 1

历史版本数 3

常见问题