remoote-mcp
Enables AI agents to search and explore remote job listings, salary stats, and popular roles, and supports token-authenticated actions like job alerts and apply redirects.
README
Remoote Remote Jobs MCP
This is the public metadata and docs repository for Remoote Remote Jobs MCP.
- Website and docs: https://remoote.app/agents
- MCP endpoint: https://api.remoote.app/remoote/agents/mcp
- Token setup: https://remoote.app/agents/connect
Remoote MCP gives agents a structured way to search current remote jobs, read job details, compare salary stats, show popular roles, create job alerts, and request user-confirmed apply redirects. It is meant for agent builders who need job data without scraping pages or exposing raw employer application links.
Tools
| Tool | Access | What it does |
|---|---|---|
search_jobs |
Public or token | Searches active Remoote jobs by query, skills, location filters, seniority, employment type, and salary visibility. |
get_job |
Public or token | Returns public details for one active job, including salary fields, location rules, skills, excerpts, and the Remoote job URL. |
get_salary_stats |
Public or token | Returns salary stats for matching jobs when enough normalized compensation data is available. |
get_top_jobs |
Public or token | Returns a capped list of currently popular remote jobs. |
create_apply_intent |
Token required | Creates a Remoote apply redirect for an authenticated user after the agent asks for confirmation. |
create_alert |
Token required | Creates a daily or weekly Remoote job alert for the authenticated user. |
Public vs. token-authenticated behavior
Public calls are available without a token for discovery and read-only job research. They are capped and return public Remoote fields, canonical Remoote URLs, and safe next actions.
Token-authenticated calls use a user-created bearer token. Use them only when the agent needs account-tied workflows such as alerts or apply intents. Create or revoke a token at https://remoote.app/agents/connect, then pass it as:
Authorization: Bearer rma_...
Only grant the scopes your workflow needs.
Apply safety
Public responses do not include raw employer apply URLs. They point users back to Remoote job pages.
create_apply_intent also returns a Remoote redirect, not a raw employer URL. The agent should ask the user to confirm before opening it, because opening the redirect may count as an application redirect or entitlement check.
Quick curl examples
All examples call the remote MCP endpoint directly over JSON-RPC.
List tools
curl -sS \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' \
https://api.remoote.app/remoote/agents/mcp
Expected result: tools includes search_jobs, get_job, get_salary_stats, get_top_jobs, create_apply_intent, and create_alert.
Search public jobs
curl -sS \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"search_jobs","arguments":{"query":"python","limit":3,"salary_required":true}}}' \
https://api.remoote.app/remoote/agents/mcp
Expected result: structuredContent.status is ok and each job links back to Remoote instead of exposing a raw employer apply URL.
Create an alert with a token
export REMOOTE_AGENT_TOKEN='rma_...'
curl -sS \
-H 'content-type: application/json' \
-H "Authorization: Bearer $REMOOTE_AGENT_TOKEN" \
-d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"create_alert","arguments":{"name":"Python remote jobs","criteria":{"query":"python","salary_required":true},"frequency":"daily","limit":10}}}' \
https://api.remoote.app/remoote/agents/mcp
Use a placeholder like rma_... in examples. Do not put real tokens in docs, logs, or prompts.
More examples
Registry metadata
server.json is prepared for MCP directories and registries. The repository URL points to the intended future public GitHub location: https://github.com/asmisha/remoote-mcp.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。