← 返回 Skills 市场
105
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install multi-search-fallback
功能描述
多源搜索聚合技能。当用户需要搜索信息、查资料、做研究时，自动调用多个搜索源进行交叉验证，提高结果准确性。触发场景：搜索某事、查证某个说法、做研究、多源验证、compare multiple sources、搜索结果不一致时主动复核。**只要是搜索类需求，一律优先使用此技能**，它会自动决定是单源快速返回还是多源深...
安全使用建议
This skill is logically consistent as a search aggregator, but it will automatically call other search skills (some of which may require API keys or access to external services) and forward user queries to them. Before installing: 1) Review the referenced downstream skills (tavily-search, deep-research-pro, mx_search, etc.) to see what credentials they require and what external endpoints they contact. 2) Decide whether you want queries (which may contain sensitive context) to be automatically forwarded to multiple external services; if not, keep this skill user-invocable only or disable global prioritization. 3) Ask the author to document which downstream tools require credentials and how data is shared/retained, and request explicit redaction rules for sensitive fields. 4) If you operate in a sensitive environment, test in a restricted sandbox or refuse installation until the transparency gaps are addressed.
功能分析
Type: OpenClaw Skill
Name: multi-search-fallback
Version: 1.0.0
The skill bundle is a search orchestrator designed to provide fallback mechanisms and cross-validation for search queries. It coordinates multiple search tools (e.g., Brave, DuckDuckGo, Tavily) using standard OpenClaw patterns like 'sessions_spawn' and contains no evidence of data exfiltration, malicious execution, or harmful prompt injection in SKILL.md or skill.json.
能力评估
Purpose & Capability
The stated purpose (aggregate multiple search sources and perform cross-validation) aligns with the skill.json and SKILL.md: it lists and intends to call multiple search skills. This dependency list is expected for an aggregator. However, the skill references tools (e.g., tavily-search, deep-research-pro, mx_search) that often require API keys or external access but the skill does not declare any required credentials or warn about them — a transparency gap.
Instruction Scope
The SKILL.md explicitly instructs the agent to spawn child sessions (sessions_spawn) and call many other skills to perform searches and cross-validate results. That is within the stated scope, but the instructions give broad discretion to automatically choose which external tools to call (including deep research and financial tools) and to send queries and results to those tools. There is no guidance about redacting sensitive input before forwarding, nor any explicit restriction on what context is forwarded to child skills. The SKILL.md also directs that this skill be prioritized for all search-class requests, which may cause unexpected automatic forwarding of user queries to external services.
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing is written to disk by the skill itself, which is low-risk from an install/execution perspective.
Credentials
The skill declares no required environment variables, yet it intends to invoke downstream skills that (according to the SKILL.md) may require API keys (e.g., Tavily). Because it auto-invokes other skills, it can cause those other skills to use their own credentials or network access without this skill declaring them. This indirect use of credentials and external endpoints is not surfaced to the user and may be disproportionate to what a user expects from a single aggregator skill.
Persistence & Privilege
The skill is not marked always:true and does not claim to persist or modify other skills' configurations. However, its autonomous invocation capability (platform default) combined with automatic prioritization for all search tasks increases the chance that user queries will be forwarded to multiple external services without explicit per-query consent. This combination raises the blast radius if downstream skills have broad privileges.
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 multi-search-fallback - 安装完成后，直接呼叫该 Skill 的名称或使用
/multi-search-fallback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 of multi-search-fallback, an aggregated multi-source search skill.
- Automatically triggers multiple search sources to cross-verify results, increasing accuracy for all search-related needs.
- Dynamically chooses between quick single-source search and deep multi-source validation based on the complexity of user's query.
- Implements a fallback search chain to ensure high availability: switches to backup sources if earlier ones fail or return empty.
- Performs cross-verification of findings: tags as "一致", "分歧", or "补充", calculates an overall confidence score, and clearly displays the consensus level.
- Structured output: provides confidence level, key findings, multi-source comparison, and collapsible raw results.
- Custom strategies for research, fact-checking, academic, finance, and domestic search scenarios.
元数据
常见问题
Multi-Search Fallback 是什么？
多源搜索聚合技能。当用户需要搜索信息、查资料、做研究时，自动调用多个搜索源进行交叉验证，提高结果准确性。触发场景：搜索某事、查证某个说法、做研究、多源验证、compare multiple sources、搜索结果不一致时主动复核。**只要是搜索类需求，一律优先使用此技能**，它会自动决定是单源快速返回还是多源深... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 105 次。
如何安装 Multi-Search Fallback？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install multi-search-fallback」即可一键安装，无需额外配置。
Multi-Search Fallback 是免费的吗？
是的，Multi-Search Fallback 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Multi-Search Fallback 支持哪些平台？
Multi-Search Fallback 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 Multi-Search Fallback？
由 ruogu1992（@ruogu1992）开发并维护，当前版本 v1.0.0。
推荐 Skills