GitHub
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
作者提供的说明(英文原文)
MCP proxy server that filters tools from upstream MCP servers via regex-based deny list
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"tool-filter": {
"command": "npx",
"args": [
"-y",
"@respawn-app/tool-filter-mcp"
]
}
}
}在终端中运行这条命令
claude mcp add tool-filter -- npx -y @respawn-app/tool-filter-mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"tool-filter": {
"command": "npx",
"args": [
"-y",
"@respawn-app/tool-filter-mcp"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"tool-filter": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@respawn-app/tool-filter-mcp"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过 GitHub 账户 respawn-llc 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(0.5.0)于6天前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:GitHub, Desktop Commander, Cloudflare MCP
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
MCP server for terminal commands, file operations, and process management
Cloudflare MCP servers
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
MCP proxy server that filters tools from upstream MCP servers via regex-based allow and deny lists.
For effective context engineering, we want to minimize useless tokens. Most major agents right now (e.g. Claude Code) do NOT remove tool descriptions from the context. Even though the tool is completely denied and unused, the model will still get its entire description and still try to call the tool (and get error messages). For big MCPs, such as github, supabase, jetbrains IDE, atlassian, this is catastrophic to the context and leads to context pollution by 40-60k of useless tokens. You wanted your agent to be able to see jira ticket descriptions? Please also have these 44 useless tools to edit assignees on confluence pages.
This MCP completely solves the issue without introducing any overhead. Contributions welcome!
# For HTTP/SSE MCP servers
npx @respawn-app/tool-filter-mcp --upstream --deny
npx @respawn-app/tool-filter-mcp --upstream-stdio --deny -- [args...]
Filter tools matching .*_file$ pattern from an HTTP MCP server:
npx @respawn-app/tool-filter-mcp \
--upstream http://localhost:3000/sse \ --deny ".*_file$"
Filter tools from a local stdio MCP server:
npx @respawn-app/tool-filter-mcp \
--upstream-stdio \ --deny "dangerous_.*" \ -- npx my-mcp-server
Pass environment variables to the upstream stdio server:
npx @respawn-app/tool-filter-mcp \
--upstream-stdio \ --env "API_KEY=secret123" \ --env "DEBUG=true" \ --deny "admin_.*" \ -- npx my-mcp-server
Use comma-separated patterns:
npx @respawn-app/tool-filter-mcp \
--upstream http://localhost:3000/sse \ --deny "get_file_text,create_new_file,replace_text"
Use --allow (-a) with comma-separated regex patterns to expose only matching tools. When both filters are set, deny takes precedence. These rules also apply to --list-tools.
npx @respawn-app/tool-filter-mcp \
--upstream http://localhost:3000/sse \ --allow "^read_file$,^search_code$" \ --list-tools --format=names
npx @respawn-app/tool-filter-mcp \ --upstream-stdio \ --allow "^read_.*,^search_.*" \ --deny "^read_secret$" \ -- npx my-mcp-server
Matching is case-sensitive and can match part of a name. Use ^ and $ to match a whole name. Blank comma-separated entries are ignored. An omitted or blank allow list leaves tools unrestricted by allow. A nonempty allow list that matches no tools exposes no tools. Invalid regex patterns and patterns rejected by the regex-safety check cause an error in proxy and list modes. Use \, to include a comma inside one pattern. For example, --allow '^[a-z]{1\,3}$' matches names of one to three lowercase letters.
Add custom headers for authentication:
npx @respawn-app/tool-filter-mcp \
--upstream http://localhost:3000/sse \ --header "Authorization: Bearer your-token-here" \ --header "X-API-Key: your-api-key"
Headers support environment variable expansion (if not yet expanded by your app):
npx @respawn-app/tool-filter-mcp \
--upstream http://localhost:3000/sse \ --header "Authorization: Bearer $AUTH_TOKEN"
Discover what tools are available before setting up filters:
# Default table format
npx @respawn-app/tool-filter-mcp \ --upstream http://localhost:3000/sse \ --list-tools
npx @respawn-app/tool-filter-mcp \ --upstream http://localhost:3000/sse \ --list-tools --format=names
npx @respawn-app/tool-filter-mcp \ --upstream http://localhost:3000/sse \ --list-tools --format=json
npx @respawn-app/tool-filter-mcp \ --upstream http://localhost:3000/sse \ --deny ".*_file$" \ --list-tools --format=names
Add to your .mcp.json:
{
"mcpServers": { "filtered-http-server": { "command": "npx", "args": [ "@respawn-app/tool-filter-mcp", "--upstream", "http://localhost:3000/sse", "--deny", "dangerous_tool_1,dangerous_tool_2" ], "type": "stdio" } } }
With authentication headers (supports environment variable expansion):
{
"mcpServers": { "filtered-http-server": { "command": "npx", "args": [ "@respawn-app/tool-filter-mcp", "--upstream", "http://localhost:3000/sse", "--header", "Authorization: Bearer ${API_TOKEN}", "--header", "X-Custom-Header: $CUSTOM_VALUE", "--deny", "sensitive_.*" ], "type": "stdio" } } }
{
"mcpServers": { "filtered-zen": { "command": "npx", "args": [ "@respawn-app/tool-filter-mcp", "--upstream-stdio", "--deny", "dangerous_.*,sensitive_.*", "--", "npx", "my-mcp-server" ], "type": "stdio" } } }
With multiple arguments to the upstream server:
{
"mcpServers": { "filtered-stdio-server": { "command": "npx", "args": [ "@respawn-app/tool-filter-mcp", "--upstream-stdio", "--deny", "admin_.*", "--", "node", "my-mcp-server.js", "--config=config.json", "--verbose" ], "type": "stdio" } } }
With arguments that start with dashes (like uvx):
{
"mcpServers": { "filtered-uvx-server": { "command": "npx", "args": [ "@respawn-app/tool-filter-mcp", "--upstream-stdio", "--deny", "test_.*", "--", "uvx", "--from", "git+https://github.com/BeehiveInnovations/zen-mcp-server.git", "zen-mcp-server" ], "type": "stdio" } } }
With environment variables for the upstream stdio server:
{
"mcpServers": { "filtered-server-with-env": { "command": "npx", "args": [ "@respawn-app/tool-filter-mcp", "--upstream-stdio", "--env", "API_KEY=${MY_API_KEY}", "--env", "DEBUG=true", "--deny", "admin_.*", "--", "npx", "my-mcp-server" ], "type": "stdio" } } }
Print the list of available tools without starting a proxy server
--list-tools: List available tools from upstream and exit--format : Output format for --list-tools (default: table)table: Human-readable table with descriptions (truncated to 100 chars)json: Full JSON output with complete schemasnames: Comma-separated tool names for easy copy-pasteTo start the proxy server, you must specify exactly one of:
--upstream : Connect to upstream HTTP/SSE MCP server--upstream-stdio: Spawn and connect to upstream stdio MCP server-- separator--upstream-stdio -- npx zen-mcp-server--allow , -a : Comma-separated regex patterns for tools to allow--deny : Comma-separated regex patterns for tools to filterOnly applicable with --upstream:
--header : Custom HTTP header to pass to upstream server (can be repeated for multiple headers)--header "Header-Name: value"$VAR or ${VAR}--header "Authorization: Bearer $TOKEN"Only applicable with --upstream-stdio:
--env : Environment variable to pass to the upstream stdio server (can be repeated for multiple variables)--env "KEY=value"--env "API_KEY=secret" --env "DEBUG=true"-- separator, provide the command and all its arguments-- is passed to the upstream server--from, --config, etc.)--upstream-stdio -- uvx --from git+https://... package-name--upstream), OR--upstream-stdio)MIT
Copyright (c) 2026 Respawn LLC