Are you the author? Sign in to claim
MCP (Model Context Protocol) server for Atmospore — pollen forecast tools for Claude and other AI assistants.
MCP server for Atmospore. Lets Claude (and other AI assistants supporting MCP) answer questions about pollen forecasts at any point on Earth.
Adds four tools to Claude:
| Tool | Use it for |
|---|---|
get_pollen(lat, lon, forecast_days) | "What's the pollen forecast in Oslo this week?" |
get_top_species(lat, lon, limit) | "What pollen is highest in Bergen right now?" |
get_area_average(lat, lon, radius_km, forecast_days) | "Is tree or grass pollen worse in London this week?" |
list_supported_species() | "What species do you cover?" |
And one resource:
atmospore://help — usage notes browsable from Claude DesktopCoordinates work anywhere on the planet — the model has global coverage at ~28 km resolution.
pip install atmospore-mcp
~/Library/Application Support/Claude/claude_desktop_config.json%APPDATA%/Claude/claude_desktop_config.json{
"mcpServers": {
"atmospore": {
"command": "atmospore-mcp",
"env": {
"ATMOSPORE_API_KEY": "ak_your_key_here"
}
}
}
}
Each tool call hits the Atmospore API and counts against your key's daily quota. The free tier (100 calls/day) covers casual use — typical Claude conversations fire 1–5 tool calls. Heavy users (~5+ pollen conversations/day) will want the paid plan.
When you hit the quota the tool returns a structured quota_exceeded response and Claude
will tell you in plain language: "You've hit your Atmospore daily limit. Upgrade at
atmospore.com/plans."
git clone https://github.com/atmospore/atmospore-mcp
cd atmospore-mcp
pip install -e ".[test]"
pytest
ATMOSPORE_API_KEY=ak_... atmospore-mcp # runs the server on stdio
MIT.
Run Claude Code as an MCP server so any agent can delegate coding tasks to it
Browser automation using accessibility snapshots instead of screenshots
Google's universal MCP server supporting PostgreSQL, MySQL, MongoDB, Redis, and 10+ databases
Official GitHub integration for repos, issues, PRs, and CI/CD workflows