Wellness Air MCP server
Local-first air-quality MCP for AI agents — AirGradient, AirThings, PurpleAir, IQAir, Awair.
0 stars
Reviews
Write oneNobody has reviewed Wellness Air yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Wellness Air tools (19, 1 write)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
air_agent_manifestReturns the wellness-air agent manifest: tool list, supported clients, env vars, recommended first calls, capabilities, privacy posture, and community links.
air_aqi_checkReturns just the AQI number + band for the configured location. Use for fast 'is the air OK to go outside?' prompts.
air_capabilitiesLists supported providers, configured providers (via env vars), available metrics, and privacy modes.
air_compare_locationsCompare current readings across multiple location IDs (default: AirGradient public sensors).
air_connection_statusReports which providers are configured, which is the default, and any warnings the agent should surface to the user.
air_current_readingFetches the most recent air-quality reading for a location. Defaults to the AirGradient public sensor at WELLNESS_AIR_DEFAULT_LOCATION (or pass locationId).
air_daily_summaryReturns a synthesized daily snapshot for a location. v0.1 returns the current reading framed as a daily snapshot; richer multi-hour aggregation arrives in v0.2.
air_data_inventoryReturns the inventory of air-quality metrics this connector exposes plus per-provider availability hints (no remote calls).
air_demoReturns a realistic example payload of what air_current_reading + air_daily_summary look like in practice. Use this to help users understand what the connector will return BEFORE they configure anything.
air_health_bandsClassify PM2.5, PM10, CO2, and VOC readings into WHO 2021 / EPA / ASHRAE / UBA health bands. Pass readings directly, or omit them and the tool will fetch the current reading from the default provider (PM2.5, PM10, CO2, VOC where available). Returns per-pollutant band + the worst signal across pollutants + deduplicated recommended actions + source citations.
air_health_recommendationQuick PM2.5-centric health recommendation. Pass current PM2.5 (µg/m³) and optionally CO2 (ppm) and VOC index; returns WHO/EPA-aligned bands per pollutant, the overall worst band, and a deduplicated list of plain-language actions. Use this when you have a reading already and just want a 'what should I do?' answer. For richer four-pollutant classification with full source citations use air_health_ba
air_list_devicesLists devices owned by the configured provider account. v0.3 supports AirThings (requires AIRTHINGS_CLIENT_ID/SECRET). For AirGradient owned sensors use air_search_public_sensors instead.
air_onboardingReturns the 11-question onboarding flow for the shared Delx Wellness profile. Read-only. The agent should ask these questions next so wellness-air (and the rest of the wellness stack) can personalize responses — non-secret data only, stored at ~/.delx-wellness/profile.json.
air_privacy_auditReturns what wellness-air stores locally, what it sends to providers, what it never logs, and agent rules.
air_profile_getReturns the shared Delx Wellness profile (~/.delx-wellness/profile.json). Read-only. Surfaces the user's preferred location, sensitivity flags (asthma, etc.), and units so wellness-air can choose tighter AQI thresholds.
air_profile_updatewrite actionPersist a partial patch to the shared Delx Wellness profile. Requires explicit_user_intent: true. Rejects any field containing oauth/token/secret/password/cookie/refresh/api_key/session — the profile is for non-secret wellness context only.
air_quickstartReturns a personalized 3-step setup walkthrough for the human based on the current agent/env state. Call this first when the user asks 'how do I use this?'
air_search_public_sensorsHelper for discovering AirGradient public sensors. Returns a curated list of well-maintained public sensors plus a hint to the AirGradient map for finding more.
air_trendWindowed trend analysis for PM2.5 / CO2 / VOC. Pulls past measurements from AirGradient (public or owned) and returns per-pollutant mean / median / min / max / current / rate_of_change_per_hour / peak_at / trough_at / time_above_threshold_minutes plus an optional natural-language observation (only when the data supports one). Pass `pollutant: 'all'` (default) for an array of per-pollutant trends p
Public scan report
scanner v0.1.10 · 2026-11-08 · same rubric, same numbers if you re-run it
- Code scan30 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancelast push 3 days ago15/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How Wellness Air compares
Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 92/100
- higher than 93% of 48,114 servers
- Tools
- 19
- the typical server has 7
- Tools that change things
- 5%
- typical: 0%. Fewer write tools means less an agent can break
What its tools let an agent do: 18 read · 1 change
Install Wellness Air in Claude Code, Cursor or VS Code
Runs npx -y wellness-air on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add wellness-air -- npx -y wellness-air
What the publisher says
From the Wellness Air repository's README, as published. We do not edit it. Read it on GitHub
<!-- delx-wellness header v2 --> Wellness Air
Local-first air-quality MCP for AI agents. Plug an AirGradient/AirThings/PurpleAir sensor into your agent — no tokens leave your machine.
📡 Why this exists: your sleep / focus / training results are partly a room-quality problem. PM2.5 above 35, CO₂ above 1,000 ppm, low humidity — those wreck recovery and the agent can't see them. wellness-air gives any AI agent local-first environmental context to cross with WHOOP / Oura / Garmin / nourish.
⚡ One-command install — pick your runtime:
- Delx Wellness for Hermes: npx -y delx-wellness-hermes setup
- Delx Wellness for OpenClaw: npx -y delx-wellness-openclaw setup
Or wire it standalone into Claude Desktop / Cursor / ChatGPT Desktop — see the install section below.
HTTP (v2 stateless)
Default is stdio. Optional Streamable HTTP — no session id, JSON responses, loopback only:
npx -y wellness-air --http
# GET http://127.0.0.1:3000/health
# POST http://127.0.0.1:3000/mcp (sessionless)Env: WELLNESSAIRHOST, WELLNESSAIRPORT, WELLNESSAIRTRANSPORT=http.
<!-- /delx-wellness header v2 -->
Overview
Wellness Air is a local MCP server that exposes air-quality readings to any MCP-aware AI agent. It ships with first-class AirGradient support (open hardware + free public API — no auth needed for the 2,000+ public sensors in the worldwide feed). AirThings and PurpleAir are implemented (bring your own free API credentials); IQAir AirVisual and Awair are on the roadmap.
If wellness-air helps your agent, please star the repo. Stars make the project easier for other AI builders to discover and help Delx keep shipping local-first wellness infrastructure.
Try It In 60 Seconds
# 89 is a real, public AirGradient sensor (Prem Tinsulanonda School, Thailand).Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Wellness Air: common questions
- Is Wellness Air MCP server safe?
- Yes, by our scan: it is graded A (92/100). Read the Wellness Air safety report
- How do I install Wellness Air?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Wellness Air need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Wellness Air maintained?
- The last commit was 10 days ago (2026-11-05). The latest release is v0.5.6.
- How does Wellness Air compare with other MCP servers?
- safety score 92/100 (higher than 93% of 48,114 servers); tools 19 (the typical server has 7); tools that change things 5% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.