直接从任何兼容 MCP 的智能体中搜索 60,000 多种 AI 智能体技能。
BytesAgain 是一个免费的 MCP 服务器，用于发现 ClawHub、LobeHub、Dify、GitHub 索引技能以及 BytesAgain 精选目录中的 AI 智能体技能和工作流用例。
该存储库包含一个真实的 stdio MCP 服务器包装器。它向智能体公开紧凑的工具，并将只读请求转发到 https://bytesagain.com/api/mcp 的公共 BytesAgain API。
Related MCP server: AgentBase MCP Server
工具 | 使用场景 | 返回内容 |
| 针对具体任务、关键词、领域或集成需求查找 AI 技能。支持英语、中文、日语、韩语、德语、法语、西班牙语和葡萄牙语。 | 包含 slug、名称、描述、类别、标签、下载量、所有者和相关性字段的排名技能摘要。 |
| 获取由 | 详细的技能元数据、安装/来源链接、类别、标签、所有者、下载量、星标以及可用时的相关字段。 |
| 当用户没有特定任务时，浏览趋势或高下载量的技能。 | 按下载量排名的热门技能。 |
| 搜索工作流/用例页面，例如“撰写周报”、“构建仪表板”或“自动化电商列表”。 | 将现实任务映射到相关技能的用例页面和描述。 |
npx -y --package github:bytesagain/mcp-server bytesagain-mcpgit clone https://github.com/bytesagain/mcp-server.git
cd mcp-server
npm install
npm startdocker build -t bytesagain-mcp .
docker run --rm -i bytesagain-mcp{
"mcpServers": {
"bytesagain": {
"command": "npx",
"args": ["-y", "--package", "github:bytesagain/mcp-server", "bytesagain-mcp"]
}
}
}变量 | 默认值 | 描述 |
|
| 公共 BytesAgain API 端点的可选覆盖。 |
无需 API 密钥。该服务器是只读的，不会向 BytesAgain、GitHub、Glama 或第三方服务写入数据。
MCP 文档: https://bytesagain.com/mcp
托管的 MCP 端点: https://bytesagain.com/api/mcp
REST 搜索示例: https://bytesagain.com/api/mcp?action=search&q=video%20editor
npm install
npm test冒烟测试通过 stdio 启动 MCP 服务器，并验证工具列表是否已正确公开。
MIT
Fetch detailed metadata for one AI skill by exact slug. Use only after search_skills or popular_skills returns a slug, or when the user provides a known slug. Do not guess slugs. Returns the skill name, description, category, tags, version, owner/author, downloads, stars, install command, source URLs, and related metadata when available. If the slug is not found, search again with related keywords instead of inventing details.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | Exact lowercase hyphen-separated slug from a previous result, e.g. "clawhub-github" or "bytesagain-video-editor". |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the burden. It implies a read-only operation by stating it 'fetches' data and returns metadata, without mentioning side effects. It could explicitly confirm non-destructiveness, but the behavior is clear enough.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured and front-loaded. It is concise but includes essential usage guidance and outcomes. Every sentence contributes meaning, though a slight reduction in length is possible without losing clarity.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (one parameter, no output schema), the description is comprehensive. It lists the returned fields and provides troubleshooting advice. It adequately covers the tool's context and user expectations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already covers the slug parameter with an example, giving a baseline of 3. The description adds value by specifying the slug must be an 'exact lowercase hyphen-separated slug from a previous result', reinforcing correct usage and validation constraints.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb 'Fetch detailed metadata for one AI skill by exact slug', clearly identifying the resource and scope. It distinguishes itself from sibling tools like search_skills and popular_skills by focusing on a single skill retrieval via slug.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly states when to use this tool ('after search_skills or popular_skills returns a slug, or when the user provides a known slug'), what not to do ('Do not guess slugs'), and provides fallback guidance ('If the slug is not found, search again with related keywords').
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Return the most popular AI agent skills by download count. Use for browsing, onboarding, trend discovery, or when the user asks what skills are popular without naming a specific task. Do not use this for targeted task matching; use search_skills for that. Returns compact skill summaries suitable for ranking lists and recommendations.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Number of popular skills to return. Default 20, maximum 50. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, but the description clearly indicates read-only behavior and what is returned ('compact skill summaries'). Could mention sorting criteria more explicitly, but overall transparent.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three focused sentences: main action, usage contexts, and exclusion. No redundant information, well-structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema, it describes the return format as 'compact skill summaries' but lacks specifics on fields. Otherwise complete for a simple list tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% for the single parameter 'limit', and the description adds no extra meaning beyond the schema's own description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('Return') and resource ('popular AI agent skills') and clearly distinguishes itself from sibling tools by stating not to use for targeted task matching.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly specifies when to use (browsing, onboarding, trend discovery) and when not to use (targeted task matching), with direct reference to alternative (search_skills).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search the BytesAgain index of 60,000+ AI agent skills by keyword or natural-language task. Use this when a user asks for tools, agents, skills, automations, integrations, or capabilities for a specific job. Supports English, Chinese, Japanese, Korean, German, French, Spanish, and Portuguese queries. Results are ranked by relevance and popularity and include slug, name, description, category, tags, downloads, owner, and score fields when available. After the user chooses a result, call get_skill with the exact slug for full details.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Number of results to return. Default 10, maximum 50. | |
| query | Yes | Search phrase or task description, e.g. "video editing", "email automation", "数据分析", or "generate product listings". |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Describes ranking, included fields, and language support. Does not explicitly state it's read-only but infers from context. Lacking annotation coverage, description carries full burden and does well but misses explicit non-mutation statement.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Concise 4-sentence description front-loaded with purpose. Every sentence adds distinct value without redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Fully explains result fields and next step. No output schema, but description sufficiently covers what to expect. Sibling tools are indirectly addressed via linkage to get_skill.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline 3. Description adds query examples and limit defaults/maximum, enhancing usability beyond schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Clearly states the verb 'Search', the specific resource 'BytesAgain index of 60,000+ AI agent skills', and distinguishes from sibling tools like get_skill (detail retrieval) and popular_skills (ranking list).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicit instructs to use when user asks for tools, agents, etc. for a specific job. Provides supported languages and a clear post-search action (call get_skill with slug).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search BytesAgain use-case pages by a real-world goal or workflow. Use this when the user describes an outcome such as "write a weekly report", "automate social media", "build BI dashboards", or asks how AI agents can help with a domain. Each result links to a use-case page with relevant skills. Combine this with search_skills when the user wants both workflow guidance and concrete tools.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Number of use cases to return. Default 10, maximum 30. | |
| query | Yes | Natural-language workflow, task, or business goal, e.g. "analyze sales data" or "write job descriptions". |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so description carries full burden. It explains results link to use-case pages with relevant skills, but lacks details on pagination, sorting, or potential side effects (though none expected for a search tool).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three sentences, front-loaded with purpose, examples, and combination guidance. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 2-param search tool with no output schema, description adequately covers input, output linkage, and usage context. Could mention result count or format, but sufficient.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with descriptions for both parameters. Description adds example queries for query param but adds no extra meaning beyond schema for limit. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states the tool searches use-case pages by real-world goal or workflow, gives concrete examples, and distinguishes from search_skills by mentioning combination for both workflow guidance and tools.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides clear when-to-use with examples and suggests combining with search_skills for broader needs, but does not explicitly exclude cases where other siblings like get_skill are more appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.1.0get_skillpopular_skillssearch_skillssearch_use_casesScored across 4 tools
Each tool serves a clearly distinct purpose: get_skill for fetching details by slug, popular_skills for browsing top skills, search_skills for keyword-based skill discovery, and search_use_cases for finding workflow-driven content. No overlap exists.
All tool names follow the verb_noun pattern using underscores (get_skill, popular_skills, search_skills, search_use_cases). While 'popular_skills' uses an adjective rather than a verb, it is consistent in style and easily understood.
With only 4 tools, the surface is lean yet sufficient for the domain of searching and retrieving AI skills and use-cases. Each tool contributes a necessary function without redundancy.
Core operations are covered: searching skills, getting skill details, listing popular skills, and searching use-cases. However, there is no tool to retrieve full details for a specific use-case, which is a minor gap given that get_skill only covers skills.
Search & install 13,000+ AI agent skills from skills-hub.ai inside any MCP tool.
Free public MCP for AI agents — 193 tools, 44 workflows. No API key.
Search and load AI agent skills. Free metadata; full content requires a Pro or Studio account.
Find AI agent skills: public previews and source links. Read-only discovery; no account required.
