Skip to content

Examples

Real walkthroughs of asking AQUAVIEW MCP questions you'd actually ask. Each folder contains:

  • The plain-English prompt
  • The transcript of the agent's tool calls
  • The result the agent returned
  • Variations to try
  • Sources used
# Folder Question Sources
01 01-quickstart/ Verify the wiring; first dataset any
02 02-sea-surface-temperature/ SST near the Florida Keys COASTWATCH, NDBC
03 03-hurricane-tracking/ Track and intensity of Hurricane Ian NDBC, GOES_R, NOAA_AOML_HDB, NOAA_GDP
04 04-tides-and-storm-surge/ CO-OPS tide gauge at Charleston, SC COOPS
05 05-argo-float-profiles/ Argo profiles around Hawaii GADR
06 06-hf-radar-surface-currents/ Surface currents off SoCal IOOS_HFRADAR
07 07-glider-missions/ Underwater glider deployments IOOS
08 08-ocean-color-chlorophyll/ Chlorophyll & HAB forecasts COASTWATCH, COASTWATCH_WC
09 09-nexrad-severe-weather/ NEXRAD radar during Hurricane Helene NEXRAD_L2, NEXRAD
10 10-hrrr-forecast/ NOAA HRRR weather model output HRRR
11 11-great-lakes/ Lake Michigan buoys near Chicago GLERL, GLOS
12 12-arctic-sea-ice/ Polar SST + sea ice from Alaska POLARWATCH, GADR, AOOS
13 13-deep-sea-exploration/ Okeanos Explorer ROV dives HYPERION, SEATUBE
14 14-vessel-traffic-ais/ AIS + MarineCadastre layers MARINECADASTRE_AIS
15 15-oil-spill-response/ Marine pollution incidents INCIDENT_NEWS, NOAA_ORR
16 16-biogeochemistry/ Cruise-based BGC + WOD profiles BCODM, WOD
17 17-bathymetry/ Hydrographic surveys, BAGs, DEMs BATHYMETRY, DIGITALCOAST
18 18-fisheries-science/ Bottom temperature & NMFS centers NEFSC, AFSC, NWFSC, PIFSC, SEFSC
19 19-regional-deep-dives/ 11 IOOS Regional Associations all IOOS RAs
20 20-international-sources/ IFREMER, EMODnet, ONC, VOTO, etc. IFREMER, EMODNET_PHYSICS, SEATUBE, ...
21 21-satellite-imagery/ Sentinel-1/2, HLS, ESA WorldCover sentinel-1-grd, sentinel-2-l2a, hls2-l30, esa-worldcover

Advanced

Deeper, technique-focused guides in advanced/:

How to run an example

  1. Make sure AQUAVIEW MCP is connected to your client — see ../INSTALL.md.
  2. Open the example's README.md.
  3. Copy the prompt into your chat.
  4. Compare what your agent does against the transcript.

The tool calls and responses shown in each transcript section were captured live against mcp.aquaview.org/mcp. The "Result the agent gave" sections are written by claude-opus-4-7 working with the catalog through the same MCP. Different models / different runs will phrase the answer differently, and the catalog itself changes over time as new datasets are indexed — so the exact item IDs, counts, and statistics you see when you copy a prompt may differ from the snapshot here.