Little Irons
About
Manage your job search directly. List jobs, create applications, add events, run company research, generate templates, and more.
Details
- Author
- Unknown
- Categories
- Productivity, Project Management, Other
Jump to
Little Irons exposes aModel Context Protocol(MCP) server that lets AI assistants manage your job search directly. List jobs, create applications, add events, run company research, generate templates, and more.
Connect your MCP client tohttps://littleirons.com/mcpusing Streamable HTTP transport (SSE).
Clients that support OAuth (like Claude Desktop) handle authentication automatically — your browser opens, you log in, and you're connected. Other clients use a Bearer token instead.
No local installation required. Everything runs through the remote endpoint.
Some clients (Claude Code, Codex, VS Code) need an API token.
You can generate one inSettingsunder the "Browser Extension" section — the same token works for both the extension and MCP.
Clients that support OAuth 2.1 (like Claude Desktop) don't need a token — they'll open your browser to authenticate automatically.
Claude Desktop supports remote MCP servers with OAuth. Add the following to your Claude Desktop config file:
- macOS:~/Library/Application Support/Claude/claude_desktop_config.json
- Windows:%APPDATA%\Claude\claude_desktop_config.json
{ "mcpServers": { "littleirons": { "url": "https://littleirons.com/mcp" } } }
Restart Claude Desktop after saving. The first time you use a Little Irons tool, Claude will open your browser to authenticate. After that, the connection persists.
Claude Code can connect to the remote HTTP endpoint directly:
claude mcp add littleirons --transport http https://littleirons.com/mcp
Or, if you prefer to use an API token directly:
claude mcp add littleirons --transport http \ -H "Authorization: Bearer YOUR_API_TOKEN" \ https://littleirons.com/mcp
Once added, Claude Code can use all Little Irons tools directly in your terminal.
OpenAI Codex supports MCP servers via the CLI. Add it with:
codex mcp add --transport http https://littleirons.com/mcp
Or pass your API token via an environment variable:
LITTLEIRONS_API_TOKEN=your_token codex mcp add \ --transport http \ -H "Authorization: Bearer $LITTLEIRONS_API_TOKEN" \ https://littleirons.com/mcp
VS Code supports MCP servers through GitHub Copilot's agent mode. Add the following to your.vscode/mcp.jsonfile (or your user settings):
{ "servers": { "littleirons": { "type": "http", "url": "https://littleirons.com/mcp", "headers": { "Authorization": "Bearer YOUR_API_TOKEN" } } } }
After adding the config, open the Copilot chat panel and switch to Agent mode. Little Irons tools will be available alongside your other MCP servers.
Any MCP-compatible client can connect to Little Irons. The general setup is:
- Endpoint:https://littleirons.com/mcp
- Transport:Streamable HTTP
- Auth:OAuth 2.1 with PKCE (auto-discovered via/.well-known/oauth-protected-resource), or pass a Bearer token in theAuthorizationheader
If your client supports a JSON config, the minimal setup is:
{ "url": "https://littleirons.com/mcp", "transport": "http", "headers": { "Authorization": "Bearer YOUR_API_TOKEN" } }
Once connected, your AI assistant has access to these tools:
- littleirons_list_jobs— List jobs with optional status filter and search
- littleirons_get_job— Get full details for a single job
- littleirons_create_job— Create a new job application
- littleirons_update_job— Update job fields
- littleirons_delete_job— Delete a job
- littleirons_update_job_status— Move a job to a new status
- littleirons_search_jobs— Full-text search across all jobs
- littleirons_add_job_event— Add an interview, call, deadline, or other event
- littleirons_research_company— AI-powered company research with web search
- littleirons_get_salary_insights— Market salary data from levels.fyi and others
- littleirons_generate_template— Generate cover letters, follow-ups, thank you notes, and negotiation emails
- littleirons_analyze_job_match— Score your resume against a job description
- littleirons_prepare_interview— Generate an interview prep guide
- littleirons_list_documents— List uploaded resumes and documents
- littleirons_get_document— Get document metadata and extracted text
- littleirons_get_user_info— View your account, plan, and usage limits
- littleirons_check_ai_quota— Check remaining AI uses for today
- littleirons_get_stats— Dashboard statistics
MCP access is unlimited. AI features (company research, salary insights, templates) use your AI quota separately.
If you're having trouble authenticating, try removing the MCP server from your client and adding it again. For Claude Desktop, you can also delete the cached OAuth tokens in your Claude config directory.
Can I use this with the browser extension?
Yes! The API token is the same one used by the browser extension. Both can be active at the same time.
MCP tools only access your own jobs and documents. The server is stateless and no data is cached between requests. All connections use HTTPS.
An MCP server for integrating with Clio practice management software, tailored for Australian legal professionals.
Multi-device file sync, dev-doc CRUD, task management, and session handoffs for AI agents - MCP + OpenAPI dual surface.
Shared workspace your AI agents write to — 184+ MCP tools for CMMN case management.
A local-first personal CRM for the terminal — manage contacts, deals, interactions, and tasks from Claude with 18 built-in tools, all stored in a local SQLite file.
Search, cite, and read PDFs from your EndNote reference library directly in Claude Desktop
Freelance business manager — clients, proposals, invoices, time tracking, scope, and follow-ups. 37 MCP tools.
A server that enables access to Joplin notes and to-dos through the Model Context Protocol (MCP).
An MCP server that uses Google's Gemini 1.5 Pro to generate concise summaries of various content types.
Connect your AI assistant to your personal knowledge base. Search, save links, create notes and to-dos. AI processes everything automatically.
Programmatic access to Google NotebookLM — via command-line interface (CLI) or Model Context Protocol (MCP) server.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.





