ReliefWeb MCP. Access global disaster and aid data in one chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
ReliefWeb MCP Server connects your AI client to the world's largest humanitarian database. It lets you search disaster reports, job postings, and organizational data from global sources like the UN and NGOs.
You can filter reports by country, theme (Health, Food, Shelter), or specific format—all using natural conversation. No API key is needed.
What your AI agents can do
Get countries
Returns country names, ISO codes, and associated disaster counts for location filtering.
Get disasters
Retrieves a list of tracked disasters, including types (earthquake, flood) and affected countries.
Get formats
Lists report formats like Situation Report or Map to filter reports by document type.
Find specific situation updates or assessments by querying filters for country, disaster type, theme, and source.
Retrieve a list of tracked disasters (earthquakes, floods, etc.), including the affected countries and general dates.
Search for available humanitarian job postings across various organizations and locations.
Restrict searches to specific report topics (like Health or Food) or document types (like Situation Report or Map).
Get a list of UN agencies, NGOs, and government bodies that publish data in the system.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
ReliefWeb MCP Server: 9 Tools for Crisis Data Management
Manage complex report searches, filter by themes or countries, find jobs, and retrieve detailed metadata using these nine structured tools.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using ReliefWeb on Vinkius019d8477get countries
Returns country names, ISO codes, and associated disaster counts for location filtering.
019d8477get disasters
Retrieves a list of tracked disasters, including types (earthquake, flood) and affected countries.
019d8477get formats
Lists report formats like Situation Report or Map to filter reports by document type.
019d8477get jobs
Searches for humanitarian job postings, providing details on organization and location.
019d8477get organizations
Retrieves names of publishing organizations (NGOs, UN) and how many reports they have published.
019d8477get report
Gets the full metadata for one specific report if you know its unique ID.
019d8477get reports
Searches and retrieves summaries of humanitarian reports using filters like date, theme, or country.
019d8477get sources
Lists the source names and types to understand where a specific report originated.
019d8477get themes
Returns topic categories (Health, Shelter) that can be used to filter reports by subject matter.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with ReliefWeb, then connect any of our 4,900+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,900+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by ReliefWeb. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 9 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Compiling crisis status reports used to mean juggling three different tabs and five separate databases.
Today, pulling together a picture of a disaster response is slow. You jump between the UN’s site for assessments, local NGOs for ground-level updates, and job boards to see who’s coming in next. You spend hours copying dates, cross-referencing country codes, and manually filtering reports by theme—it's tedious, error-prone work.
With this MCP server, you simply tell your agent what you need: 'What is the status of shelter provision in Haiti?' The system runs `get_reports` across multiple sources, uses `get_themes` to isolate Shelter data, and returns a single, clean summary. It cuts out all the clicking.
ReliefWeb MCP Server gives you instant access to global humanitarian reports.
Before this tool, finding specific job openings meant checking multiple NGO career portals individually. If you were looking for a WASH Coordinator role in South Sudan, you'd have to visit dozens of sites and sift through thousands of listings.
Now, you ask the agent to find it. It uses `get_jobs` instantly, filtering by both type and location across the entire indexed database. You get actionable results—not links to dead ends.
What you can do with this MCP connector
The get_reports tool lets your AI client search and pull summaries of humanitarian reports using filters for date, country, theme, or source.
You can use get_themes to narrow down searches by topic category like Health or Shelter. You'll also find the get_formats tool, letting you restrict results to specific document types such as Situation Reports or Maps. To pinpoint exactly where a crisis is hitting, your agent runs get_countries, which returns country names, ISO codes, and how many disaster counts are tied to that location.
If you need to understand the full scope of a situation, the get_sources tool lists every source name and type, showing you exactly where a specific report originated. When you're looking for reports on a particular event, running get_disasters retrieves a list of tracked disasters—like earthquakes or floods—and shows which countries were affected.
To deep-dive into one specific piece of information, the get_report tool grabs all the metadata for that single report if you know its unique ID. You can also use get_organizations to pull a list of publishing bodies—NGOs, UN agencies, or government groups—and see how many reports each organization has put out.
Finding job openings in aid work is simple; the get_jobs tool searches for active humanitarian postings, providing details on both the hiring organization and the location. You can also get a clear picture of who's publishing data overall by using get_organizations to list all the major players involved.
It’s really straightforward: your AI client just talks to ReliefWeb. It pulls global information that covers everything from assessing damage after a flood or an earthquake, to finding job openings with international groups. The system tracks crises and locations through structured data retrieval for you. You'll find summaries of reports by querying filters for country, disaster type, theme, and source name.
Your agent can also get the full list of available topics using get_themes, making it easy to restrict searches to specific subject matter like Food or Water Security. It’s a massive database, so knowing how to filter is everything.
019d8477-2ec5-70b1-854c-1e76f04574f2 
How ReliefWeb MCP Works
- 1 Subscribe to this server and connect your AI client. No API keys are needed for setup.
- 2 Ask your agent a natural language question (e.g., 'Show me health reports from Yemen').
- 3 The system runs the necessary tools, compiles the data, and returns structured results.
The bottom line is you get direct access to global, filtered humanitarian data without manually navigating multiple websites or building complex API calls.
Who Is ReliefWeb MCP For?
This is for people who need quick answers from massive amounts of public-facing data. It helps journalists write faster during a crisis and researchers track trends across decades. If your job involves compiling status reports or finding operational guidance, this server saves hours.
Needs to quickly find the most recent situation report for their specific region after a crisis event.
Studies long-term trends by comparing health reports across different countries and time periods over years.
Gathers multiple sources, job listings, and impact assessments from varied organizations to build a comprehensive story.
What Changes When You Connect
- Find the right report fast. Instead of browsing through hundreds of PDFs, you ask your agent to use
get_reportswith filters for 'country: Sudan' and 'theme: Health'. You get a curated list immediately. - Track crises across time. Use
get_disastersfirst to see what disasters are active or recently occurred in a region. Then, you runget_reportsto find the official assessments on those events. - Keep your team staffed. The
get_jobstool lets you search for specific roles (e.g., 'WASH Coordinator') across multiple NGOs worldwide without leaving your chat window. - Understand data depth. If a report is vague, use
get_sourcesto confirm who published it and check how many total reports that organization has filed viaget_organizations. - Filter complex data sets. You don't just search by keyword; you narrow the results using
get_themes(e.g., 'Protection') orget_formats(e.g., 'Assessment').
Real-World Use Cases
Assessing a recent earthquake's impact.
The field worker needs to know the damage scope. They ask their agent: 'Find all reports on earthquakes in Nepal from 2015 and 2023.' The agent uses get_disasters first, then runs get_reports, filtering by date and disaster type. This gives them a quick overview of historical responses.
Staffing an emergency deployment.
A coordinator needs to fill several roles quickly. They ask the agent to 'Show me protection officer jobs in Kenya.' The agent uses get_jobs, filtering by role and location, providing immediate links and deadlines.
Writing a cross-sectoral article.
A journalist needs context on multiple issues. They ask the agent to 'Gather reports covering both food security and shelter in Ethiopia.' The agent uses get_themes to confirm the IDs, then runs get_reports twice (once for each theme) to compile a comprehensive draft.
Building an organizational map.
A researcher wants to see who is active in a region. They ask: 'Which UN agencies and NGOs have published reports on clean water?' The agent uses get_organizations first, then runs get_reports, filtering by topic (Health/Water) to build the list.
The Tradeoffs
Treating it like a general search engine
Asking 'What happened in Africa?' is too vague. The agent doesn't know if you mean health, disaster, or jobs.
→
Be specific. First, use get_countries to narrow the region down (e.g., 'Nigeria'). Then, specify what type of data you want: 'Run get_reports for Nigeria, filtering by theme: Health.' This uses structured filters.
Ignoring report types
Assuming all reports are written in the same way. You might miss a critical map or assessment because it wasn't filed as a standard 'report'.
→
Always check get_formats first. If you need geographical context, ask for 'Map' formats specifically, even if your main query is about floods.
Trying to get one report by memory
Asking the agent to 'Find that big report I saw last year on Yemen.' You don't know the ID and the AI can't guess it.
→
Use get_reports with as many known filters as possible—date range, country, theme. If you remember an organization, use get_organizations first.
When It Fits, When It Doesn't
You should use this server if your need is synthesis: compiling data from multiple sources (disasters, jobs, reports) or needing highly structured filtering based on defined categories (country IDs via get_countries, themes via get_themes). It’s best for building a narrative picture of a crisis.
Don't use it if you just need to know one piece of information from one source. For example, if you only want the latest job posting in one city, using get_jobs is fine. But if your goal was simply to query a live database record that doesn't fit into these structured categories (e.g., real-time stock prices or weather data), this server won't help—you’d need a dedicated API client for financial or meteorological services instead.
Common Questions About ReliefWeb MCP
How do I use get_reports to filter for a specific disaster? +
You include 'disaster' in your prompt, or you can first run get_disasters to confirm the exact name (e.g., 'Cyclones'). Then, ask the agent to search reports using that specific event name.
Do I need an API key for get_jobs? +
No. You don't need any keys. Just connect your AI client and ask the agent to find jobs; it handles the connection through the MCP standard.
What is the best way to use get_themes? +
Use get_themes first to see all available topics (like Food, Protection, Health). Then, include that theme name in your search query for get_reports.
Can I find reports about a specific organization using get_organizations? +
Yes. You can use get_organizations to check the full list of publishing bodies and see how many reports each one has submitted, giving you an idea of their activity level.
How do I use get_countries to get an ID for filtering with get_reports? +
You must run get_countries first. This tool returns a list of all country names and their corresponding ISO codes, which you need to pass into get_reports to filter results accurately.
What specific data points can I retrieve using the get_report command? +
get_report pulls full metadata for one document. You get the title, body content, source, associated themes, countries, disasters, and any attached files all in one call.
If I need to filter reports by type, how do I check valid formats using get_formats? +
Run get_formats. It lists every available report format—like Situation Report or Assessment—along with their unique IDs. Use these IDs in subsequent searches for precise filtering.
Does get_disasters list all possible disaster types, or just currently tracked ones? +
get_disasters returns the names and categories of disasters (e.g., earthquake, flood, cyclone). This gives you the full vocabulary needed to search for reports across any type.
Do I need an API key? +
No! ReliefWeb's API is completely free and open. No authentication required. Just subscribe and start searching. Data is updated continuously.
What kind of reports are available? +
ReliefWeb hosts 300,000+ reports including situation reports, assessments, press releases, maps, infographics and evaluations from UN agencies, NGOs, governments and research organizations.
Can I search reports by country? +
Yes! Use get_reports with the country parameter to filter reports by country name. You can also filter by disaster type, theme, format, date range and organization simultaneously.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.