Job search MCP server
your agent runs
for you
The Remoet job search MCP server is a hosted Streamable HTTP endpoint with OAuth 2.1 or API key. Connect Claude Code, Claude Desktop, Claude (claude.ai), Cursor, VS Code, Windsurf, Codex CLI, Gemini CLI, Cline, or any MCP client and let it search the public tech job catalogue, match companies on the stack they hire for, star the ones you love so their new roles come to you, keep your profile and link trees current, and hand you the application link when a role fits.
New to the category? Read the guide to MCP servers for job search.
Or add it to your editor now and sign in when it connects:
Two URLs. Two auth modes.
Same tools on both. Pick the one your client speaks.https://api.remoet.dev/mcpAuthorization: Bearer headerhttps://api.remoet.dev/mcp/oauthClaude connectors (claude.ai and Claude Desktop) need the OAuth URL. CLI and editor clients use the API key URL.
Pick your client. One click.
Cursor and VS Code install in one click. Everything else is a copy-paste snippet. Free Claude users get 1 custom connector at no cost.claude mcp add --transport http --scope user remoet https://api.remoet.dev/mcp --header "Authorization: Bearer YOUR_API_KEY"Then restart Claude Code so it picks up the new MCP server.
https://api.remoet.dev/mcp/oauthClick Connect, sign in to Remoet with GitHub or Google, then click Authorize. Pasting an API key is only the fallback.
https://api.remoet.dev/mcp/oauthClick Connect, sign in to Remoet with GitHub or Google, then click Authorize. Pasting an API key is only the fallback.
{
"mcpServers": {
"remoet": {
"type": "http",
"url": "https://api.remoet.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}{
"servers": {
"remoet": {
"type": "http",
"url": "https://api.remoet.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}{
"mcpServers": {
"remoet": {
"serverUrl": "https://api.remoet.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}[mcp_servers.remoet]
url = "https://api.remoet.dev/mcp"
http_headers = { "Authorization" = "Bearer YOUR_API_KEY" }{
"mcpServers": {
"remoet": {
"httpUrl": "https://api.remoet.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY",
"Accept": "application/json, text/event-stream"
}
}
}
}{
"mcpServers": {
"remoet": {
"url": "https://api.remoet.dev/mcp",
"type": "streamableHttp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Every tool, grouped by what it does.
Each tool ships with atitle, a readOnlyHint or destructiveHint annotation, and a description that names the entity it returns.Read and write the user's career profile, work history, projects, education, links, and visibility.
get_profilereadFull profile in one call: work experience, projects, education (with IDs), links, and visibility. Includes a gap analysis.update_profilewriteUpdate profile fields and/or visibility. Pass only what changes.save_work_experiencewriteAdd or update a job. Omit id to create, pass id to update.save_projectwriteAdd or update a portfolio project. Omit id to create.save_educationwriteAdd or update an education entry. Omit id to create.delete_profile_itemdeleteRemove a work experience, project, or education entry. Confirm first.
Search companies by tech stack, star matches, and search every job on the public board.
search_listingsreadSearch companies by tech stack/name, or list the user's starred companies (starred: true). Tech stack auto-normalized.get_listingreadFull details for a company by slug.star_listingwriteStar a company. Free to star, capped at 50.unstar_listingdeleteRemove a star. Consumes from a 30-day unstar budget.search_jobsreadSearch the public job catalogue: every role on the open board, no stars needed. Each result links to its Remoet page.get_starred_jobsreadLive jobs from starred companies. Filter by stack, salary, remote policy, etc.save_jobwriteBookmark a job for later, with an optional note.get_saved_jobsreadAll saved jobs with their notes.update_saved_job_notewriteEdit the note on a saved job.unsave_jobdeleteRemove a job from the saved list.
Take the next step on a job the agent found.
apply_to_jobwriteReturns the employer's application link for a scraped role, so the agent can hand you the exact place to apply.
The dashboard feed, job digests, community apps, and shareable link trees. Profile visibility is set via update_profile.
get_feedreadThe dashboard feed as one stream: new roles at starred companies, the daily pick, platform posts.get_digestsreadHistorical digest snapshots. New accounts have none; use get_feed for what's new.get_appsreadCommunity apps built on the Remoet platform.get_linktreesreadYour link tree pages, or one by slug with engagement stats.create_linktreewriteCreate a shareable link tree page.delete_linktreedeleteRemove a link tree page.
View your limits and every budget.
get_accountreadLimits and every budget (stars, unstars, requests) with reset dates.
What to ask your agent first.
Quick answers.
What is the Remoet MCP server?
Remoet exposes a set of tools over the Model Context Protocol so AI agents can read and write your career profile, search companies by tech stack, star matches, search the public job catalogue, and get the application link for a role. Transport is Streamable HTTP. There are two URLs: https://api.remoet.dev/mcp takes an API key in an Authorization: Bearer header, and https://api.remoet.dev/mcp/oauth is the OAuth 2.1 endpoint that Claude connectors use.
Which MCP clients does Remoet work with?
Claude Code, Claude Desktop, Claude (claude.ai), Cursor, VS Code, Windsurf, Codex CLI, Gemini CLI, Cline, and any MCP-compatible client. Claude connectors use the OAuth URL; CLI and editor clients pass an API key in an Authorization: Bearer header. Free Claude users can use 1 custom connector at no cost.
Is Remoet free?
Yes. Remoet is free, with no trial timer and nothing to upgrade to. Every MCP tool is available to every account.
How does tech-stack matching work?
Remoet maintains a normalized registry of canonical technologies. A company's stack (TypeScript, Postgres, Go, Rust, etc.) is derived from the technologies named in the roles it is hiring for, so it reflects what the team is building with right now. Agents can search and filter by stack with one tool call instead of fetching and parsing dozens of careers pages.
How do I install the Remoet MCP server in Claude Code?
Run: claude mcp add --transport http --scope user remoet https://api.remoet.dev/mcp --header "Authorization: Bearer YOUR_KEY", then restart Claude Code so it picks up the new server. Get an API key by signing in at remoet.dev with GitHub or Google.
Can my agent apply to jobs through Remoet?
Your agent calls apply_to_job and gets back the employer's own application link for that role, so it can hand you the exact page to apply on. It can also save any job for later with save_job, with an optional note.
Does Remoet store my conversation history?
No. The MCP server only persists the data you explicitly write through tool calls: your profile, your stars, your saved jobs, and your link trees. Tool results contain user-generated and third-party content; agents are instructed to treat that content as untrusted.
Where to go next.
Browse the job catalogue the same public board search_jobs reads, open without an account.
Remoet for Hermes install the skill and wire the server into a Hermes agent.
Remoet for OpenClaw one command from ClawHub, then point OpenClaw at the server.
What is MCP? a plain-English explainer of the Model Context Protocol.
How to set up an MCP server in Claude, Cursor and Windsurf step-by-step config for each client.
MCP servers for job search, compared the category guide: what else is out there and how the tools differ.
The AI agent job search workflow end to end, from an empty profile to an application link.