Search, preview, and measure Umbra open SAR from any MCP client. Umbra ships no search API.
Umbra MCP Server (io.github.reesehammer/umbra-mcp)
This MCP server lets any MCP client search, preview, and measure Umbra open SAR. According to the source, Umbra ships no search API, so the server provides these capabilities directly to MCP clients. The implementation is associated with the umbra-py project.
🛠️ Key Features
Search Umbra open SAR from MCP clients
Preview results
Measure SAR outputs
No Umbra search API is provided (server provides search capability)
🚀 Use Cases
Querying Umbra open SAR datasets via an MCP client
Reviewing SAR content before processing
Extracting measurements from SAR data
⚡ Developer Benefits
Works with “any MCP client” via Model Context Protocol
Uses the umbra-py codebase (Python 3.10+), with Apache 2.0 licensing
Project provides CI and coverage badges, plus documentation at umbra-py.space
⚠️ Limitations
Search relies on the server since Umbra “ships no search API”
Topics
earth-observationsar
Captured live from the server via tools/list.
search_catalog
Search Umbra's catalog and return compact context cards.
Filters (all optional, combine freely): ``bbox`` as
``[min_lon, min_lat, max_lon, max_lat]`` in WGS84 degrees; ``intersects`` as
a GeoJSON polygon object (a ``Polygon`` / ``MultiPolygon`` geometry, or a
``Feature`` / ``FeatureCollection`` wrapping one) to keep only footprints
intersecting it -- a tighter spatial filter than the rectangular ``bbox``,
and mutually exclusive with ``bbox`` / ``place``; ``place`` as a
free-text name geocoded to a bbox; ``area`` as a substring of the Umbra
task (site) name (set ``fuzzy`` to match it loosely -- word-order- and
punctuation-independent and typo-tolerant, resolved deterministically with
no model call, so ``"beet piler"`` still reaches ``"Beet Piler - ND"`` and
``"utah centerfield"`` still reaches ``"Centerfield,
Utah"``; set ``semantic`` instead to treat ``area`` as a plain-language
*description* of a site whose name you don't know and resolve it to the
closest task names by *meaning* via the prebuilt embedding index, so
``"grain storage north dakota"`` reaches ``"Beet Piler - ND"`` -- a query
that shares no word with the label, which ``fuzzy`` cannot and should not
fake; ``semantic`` and ``fuzzy`` are mutually exclusive, and the resolved
task name is returned as ``resolved_area`` with the ranked candidates in
``semantic_matches`` so you can audit and retry with a different one, and
``min_score`` -- a cosine threshold in ``[0, 1]`` -- drops weak aliases, so
a description that matches nothing confidently returns an empty result
rather than an arbitrary top pick); ``start``/``end`` as acquisition-date
bounds -- an ISO
``YYYY-MM-DD`` date, a bare year/month (``2024``, ``2024-03``), or a
relative expression (``today``, ``yesterday``, ``3 months ago``,
``last month``), resolved deterministically with no model call;
``products`` to restrict to product types
(any of GEC, SICD, SIDD, CPHD); the SAR acquisition-property filters
``polarizations`` (keep items exposing any of these, e.g. ``["VV"]`` -- the
filter that keeps a change comparison like-with-like), ``min_incidence`` /
``max_incidence`` (view incidence-angle bounds in degrees) and
``max_resolution`` (keep items at least this fine, in metres) -- each a hard
predicate that excludes items missing that metadata; ``canonical`` to keep
the preferred copy of a collect Umbra republished under another prefix
(``True``), only the other copies (``False``), or every item when omitted;
``limit`` to cap
results; ``max_per_task`` to cap items per site (use 1 for a one-pin-per-site
overview).
``local`` selects the backend: leave it unset to use the on-disk index when
present (instant) and fall back to a live S3 walk otherwise. When the server
is configured with a Canopy token (``$UMBRA_CANOPY_TOKEN``), the search runs
against Umbra's authenticated commercial archive instead and ``source`` is
reported as ``"canopy-archive"``. Returns the per-item context cards, not
full STAC JSON, to protect the context window; call ``get_item`` for one
item's full metadata.
Prefer ``fuzzy=True`` (or :func:`find_repeat_sites`) when ``area`` shares
words with a task name; that path is deterministic and needs no key.
``semantic=True`` is only for a description that shares **no** tokens with
the label. It requires a task-embedding index built ahead of time with
``umbra semantic build`` and a user-supplied embedding key (``OPENAI_API_KEY``,
optionally ``OPENAI_BASE_URL`` / ``UMBRA_EMBED_MODEL``); it raises a helpful
setup error otherwise, so it never runs implicitly. The only model call is
turning ``area`` into a vector -- the resolved search over the chosen backend
stays fully deterministic.
Rank the archive's most repeat-imaged sites — where change detection has
something to measure.
The discovery step *before* every analysis verb. ``change`` / ``timescan`` /
``stack`` / ``pick_change_interval`` / ``narrate_change`` all answer *what*
changed at a site, but each assumes the caller already knows *which* site to
point them at. Umbra files every pass of an area under one task (site), so a
site's coverage is just how many dated passes share its task; the sites with
the most are exactly the ones where a time series exists to analyse. This
tool ranks them, so ``find_repeat_sites → pick_change_interval →
narrate_change`` is a complete chain a model can drive with no site known in
advance — the deterministic answer to "where is there something to look at?"
(the discovery moat; no model is called). Reach for this
(or ``search_catalog(area=..., fuzzy=True)``) before ``semantic=True`` when
the site name is roughly known.
The filters scope the *pool* of acquisitions the ranking is computed over,
and are the same ones ``search_catalog`` takes: ``bbox``
(``[min_lon, min_lat, max_lon, max_lat]`` WGS84), ``intersects`` (a GeoJSON
polygon, mutually exclusive with ``bbox`` / ``place``), ``place`` (free-text,
geocoded to a bbox), ``area`` (a task-name substring, loosened by ``fuzzy``),
``start`` / ``end`` date bounds, ``products`` (GEC / SICD / SIDD / CPHD), and
the SAR property filters ``polarizations`` / ``min_incidence`` /
``max_incidence`` / ``max_resolution``. ``top`` caps how many sites are
returned, and ``min_passes`` is how many passes a site needs to qualify (2 is
the minimum a change composite can use; raise it to find only deeply-revisited
series), counted on the depth ``rank_by`` measures -- raw dated passes by
default, the usable (comparable) series' depth under ``rank_by="comparable"``.
``rank_by`` chooses the order: ``"passes"`` (the default) ranks by raw pass
count; ``"comparable"`` ranks by *analysable* depth instead -- the
``comparable_passes`` largest same-polarization dated subset a change verb can
actually difference -- so a deep single-polarization series is not outranked by
a broader mixed-polarization site whose passes no analysis verb can compare, and
``min_passes`` then floors that same usable depth, so ``rank_by="comparable",
min_passes=3`` returns only sites whose differenceable series is at least three
passes deep. The two agree when every dated pass of every site shares one
polarization; prefer ``"comparable"`` when the next step is
``pick_change_interval`` / ``stack_stats``, which consume exactly that
differenceable subset. Three *temporal* orders rank on the axes the filters below
select on: ``"recency"`` orders by each site's newest dated pass (the
still-active site to monitor or task, which a depth order buries under a deeper
but dormant series), ``"span"`` by each site's observation baseline (the site
watched long enough for slow change to show), and ``"cadence"`` by each site's
*typical* revisit gap, tightest first (the most-frequently-imaged site -- the
median gap, not the worst one, so a single outage does not bury an otherwise
reliably-imaged series; the worst-case reading stays the ``max_revisit_days``
filter). ``min_passes`` still qualifies a site on raw depth under any of the
three, so a temporal order returns only sites with a series worth ordering.
``active_since`` keeps only sites still imaged *on or after* that date -- a
recency filter on each site's newest pass, so a deep series that stopped long
ago is dropped and an actively-revisited one is kept. Use it to find live
monitoring targets ("repeat-imaged sites still being collected"). It accepts an
ISO date (``"2024-06-01"``), a bare year/month, or a relative expression
(``"6 months ago"``), is orthogonal to ``rank_by`` / ``min_passes`` (it gates on
the site's latest pass, whatever depth the ranking measures), and differs f
Return the full context card for one item.
``url`` is the JSON URL of an open-data STAC sidecar, a GEC/SICD (or other
product) filename such as ``2025-05-17-17-43-46_UMBRA-08_GEC.tif``, or the
acquisition stem ``2025-05-17-17-43-46_UMBRA-08``. A filename or stem is
resolved through the local catalog index: the stem is matched against
asset hrefs, which is what distinguishes ``sar-data/tasks/``,
``sar-data/task-data/``, and ``open-data/`` — the name itself does not.
Zero matches is an error. More than one match lists the candidate STAC
URLs; pass one of those. The hosted server resolves a filename or stem
with the catalog index it already loaded. A local server does the same
after ``umbra index fetch``.
When the index contains a STAC URL, the indexed snapshot is returned —
the same document ``search_catalog`` already answered from — rather than
a live re-fetch that can disagree on asset filenames or product types.
When the server is configured with a token (``$UMBRA_CANOPY_TOKEN``) and
``url`` is a bare acquisition id (no ``://``, and not a filename or stem),
it is instead looked up in Umbra's authenticated commercial archive by a
keyed STAC search — the retrieval complement to ``search_catalog`` over
the paid catalog, since archive cards carry an ``id`` (not a fetchable
sidecar URL) as their hand-off pointer. A filename or stem still resolves
from the local open-data index when a token is set.
On the public host a URL is fetched only when it is https on Umbra's
public open-data bucket. Any other URL is refused before a request.
A bare id that is not in the local index is refused the same way.
The card (:meth:`UmbraItem.to_llm_context`) carries the id, ISO datetime,
place, bbox, resolution, polarization *with the change-detection caveat*,
the per-product-type explanations and asset URLs, and the mandatory
attribution line — everything a model needs to reason about the scene.
Resolve a free-text place name to a bounding box via Nominatim.
Returns ``{"bbox": [min_lon, min_lat, max_lon, max_lat], "display_name":
...}`` so the agent can inspect the box before searching, or pass it
straight to ``search_catalog``'s ``bbox``.
Report on the local catalog index, if one has been built or fetched.
Lets a long-running agent decide whether searches will be instant (index
present) or pay a live S3 walk, and how stale the snapshot is. ``end`` is
the newest acquisition that is not ``datetime_suspect``.
``datetime_suspect_count`` is how many kept rows carry that flag.
``quarantined_future_count`` is how many source rows the last build
dropped for a datetime more than a day ahead (they are not in the index).
``unique_captures`` is how many rows are the canonical copy of their
collect; it equals ``items`` until republished copies have been linked.
Return a scene's quicklook PNG as an image block.
Prefers a baked catalog preview (``preview="auto"``, the default) so a
public host can show the radar scene without proxying Umbra COGs.
**A baked preview is typically 512 px** (the published weekly sidecar).
``max_size`` applies only to a fresh ``preview="render"`` stream; when a
cached thumbnail is smaller than requested, the caption says
``max_size`` was ignored *and* includes a JSON
``{"image": {source, width, height, requested_max_size, substituted}}``
sidecar. For a higher-resolution picture (typically 1024 px), run a
local ``uvx --from 'umbra-py[mcp]' umbra-mcp`` server, or download the
GEC href from ``get_item``. ``preview="baked"`` refuses rather than
stream; ``preview="render"`` always streams (needs ``[viz]``, refused
on ``umbra serve --public``).
Composite 2-3 passes of one site into a change image.
Pass the STAC URLs of two or three acquisitions of the *same* site in time
order. Colors encode change: bright green = signal appeared after the first
date, magenta = signal vanished, grey = unchanged. Refuses to mix
polarizations (HH vs VV are not comparable). Returns the image block plus a
caption naming the color semantics and attribution -- and, when a pass carries
no polarization metadata so same-polarization could not be *verified*, a
structured caution text block naming that gap.
Summarize a whole time series into one activity image.
Pass the STAC URLs of a site's passes; the timescan encodes per-pixel
temporal statistics so bright/colored areas are *where activity happened*
over the series. Returns the image block plus attribution.
Measure how much a site changed across its passes, as numbers.
The quantitative counterpart to ``change_composite`` / ``timescan``: those
return a picture of *where* change is, this returns the *how much*. Pass the
STAC URLs of a site's passes (2 or more, any order); they are co-registered
onto one shared grid and reduced to a per-pass statistics series plus a net
baseline-to-latest record — mean/median/spread of each pass, the signed
decibel change against the pass before it, the fraction of ground that moved
past ``change_threshold_db``, and that fraction as an **area in km²**.
Change is always reported in decibels (a ratio of backscatter is a difference
on the log scale). The default ``crs="utm"`` puts the grid in the site's UTM
zone so every cell covers the same ground and the area figures mean
something; pass ``crs="EPSG:4326"`` for a lon/lat grid, and the areas come
back ``None`` rather than wrong. ``extent="intersection"`` (the default)
compares only ground every pass covers; ``"union"`` keeps all of it and
compares per cell where two passes overlap. ``bbox`` (lon/lat
``[min_lon, min_lat, max_lon, max_lat]``) narrows the measurement to a
sub-area, e.g. one facility inside the scene.
``blocks=N`` adds the *where* to the *how much*: the scene is cut into an
N x N grid (6 is a good default; 0, the default, skips it) and the extra
``spatial`` key reports each block's own net change, a compass label and
lon/lat centre to locate it by, and the consecutive pair of passes that
block moved most between — plus ``peak_block`` and an ASCII ``grid_text``
heat-map of the net change. Ask for it whenever the question is *where on
this site* something happened, or when a scene-wide mean looks small but the
picture from ``change_composite`` shows an obvious hot spot: a change
confined to one corner is averaged away scene-wide and stands out per block.
``block_series=True`` (which needs ``blocks``) keeps each block's *whole*
pass-to-pass sequence rather than only its peak interval, so a block that
drifted a little every pass reads differently from one that jumped once and
held. It is the largest payload this tool emits — ``blocks`` × ``blocks`` ×
(passes − 1) records — so ask for it only when the *shape* of the history is
the question, not the size of the change.
``speckle_filter="boxcar"`` or ``"lee"`` averages **speckle** down before
anything is measured. Speckle is not sensor noise: it is the interference
pattern coherent illumination makes on rough ground, and on single-look
imagery it scatters as widely as its own mean — so on a quiet site an
unfiltered per-cell decibel delta is mostly interference rather than change.
``"boxcar"`` averages every window (most variance removed, blind to the edges
it averages across); ``"lee"`` averages only where a window is no more
variable than speckle alone explains, so edges and bright points survive.
Reach for it when the change you are looking for is small, or when the
numbers look noisy against a picture that looks quiet; leave it off when the
question is about *fine* structure, because what it spends is resolution — a
``speckle_window``-cell window (odd, 5 by default) makes each cell report
ground that many cells across. The returned ``caveats`` say so, so a quoted
figure carries the trade with it.
Refuses to mix polarizations (HH and VV are not comparable). **No model is
called** — this is arithmetic over streamed COG overviews, so the numbers are
reproducible and safe to quote. Read the ``caveats`` before interpreting
them: Umbra's open products are not radiometrically calibrated, so a decibel
value is relative to the same ground on another date, and look geometry moves
backscatter too.
Check whether a site's passes are one measurement, before measuring them.
``stack_stats`` refuses a selection whose rasters were made by *different*
``umbra convert`` settings — a calibrated pass differenced against an
uncalibrated one puts the difference between two conversions on the time
axis and reports it as change on the ground. This is the question that
refusal answers, asked ahead of it: it reads each acquisition's ``UMBRA_*``
conversion record straight from its raster header (kilobytes per pass, no
pixels, no co-registration) and groups the selection by what its pixel
values *are*.
Returns ``agrees`` (whether ``stack_stats`` would accept the selection),
``groups`` largest-first — each with the conversion ``record`` its members
share, their ``item_ids`` and the ``hrefs`` to re-run on — plus ``shared``
(the record they all agree on) when they agree, ``refusal`` (verbatim the
error ``stack_stats`` would raise) when they do not, and ``unreadable`` for
any source that could not be opened at all, which is an incomplete answer
rather than a mixed one.
Reach for it in two situations. **After** a ``stack_stats`` call fails with
a provenance refusal: ``groups[0]["hrefs"]`` is the largest set of passes
that *do* agree, so re-running ``stack_stats`` on those is a measurement
rather than a guess — the refusal itself cannot name that subset. And
**before** measuring a long series you did not assemble yourself, since the
check costs a header read per pass and saves the whole cube when the answer
is no.
A series of Umbra's published products as they ship (the common case) agrees
trivially: every pass reads as ``"(unrecorded)"``, which is one value, so
``agrees`` is true and ``shared`` is empty. **No model is called.**
Scan a site's whole series and name the two passes worth narrating.
The deterministic first half of the "watch a site over time" story, and the
step that turns a wall of passes into an answer. Given the STAC URLs of a
site's passes (2 or more, any order), it co-registers them into a
``to_stack`` cube, reduces it with ``stack_stats``, and returns the *one*
consecutive pass-pair whose measured change stands **furthest clear of the
speckle detection floor** — the pair a reader should look at first. No model
is called: a number picks the frames (the determinism boundary applied to
frame selection), so the choice is reproducible and safe
to quote.
Why this exists as its own tool: ``narrate_change`` reads a *pair* of passes,
but a site has many, and "which two?" is a measurement question, not a model
one — a picture past three dates encodes nothing a model can separate, so the
pair has to be chosen before the model is called. This answers it. The
returned ``urls`` are the two chosen passes in time order, ready to hand
straight to ``narrate_change`` (or ``change_composite`` to see them): the
two tools chain into scan → narrate with no server and no model in the
selection.
Returns ``selected_interval`` (the interval's ``earlier_id`` / ``later_id``,
their datetimes, the ``changed_fraction`` and signed ``mean_delta_db``, the
cube's ``false_alarm_fraction`` speckle floor, the ``excess`` of the change
over that floor, and ``stands_clear`` — whether the change cleared the floor
by a usable margin), ``urls`` (the earlier then the later pass's STAC URL),
and the license/attribution. When ``stands_clear`` is false the series' own
largest change is still inside the speckle, so the pair is offered but the
reader is warned the difference may be interference rather than change.
``selected_interval`` is ``None`` (and ``urls`` empty) when the series has
fewer than two comparable passes or no interval reported any change — there
is no pair to narrate. ``crs="utm"`` (the default) grids the cube in the
site's UTM zone so the change fractions weigh equal ground; ``max_size`` and
``change_threshold_db`` match ``stack_stats``. Refuses to mix polarizations
(HH vs VV are not comparable). Requires the ``load`` extra.
Download one asset of an item to local disk, gated by a size check.
SAR products can be multiple GB, so this is a two-step handshake: call it
first with ``confirm=False`` (the default) to get the byte size without
downloading, then again with ``confirm=True`` to fetch it. On a
local or stdio server, ``confirm=True`` returns ``{asset, path, bytes}``.
``confirm=False`` returns a size-and-hint object and does not write a file.
On a public host (``umbra serve --public``, COG streaming off),
``confirm=True`` does not transfer the file. It returns a link plan:
``mode`` ``"link"``, ``downloaded`` false, ``asset``, ``filename`` (a
basename), the direct public ``href``, ``bytes``, and ``how_to_fetch``.
When the href is https on the public open-data bucket, ``how_to_fetch``
is ``curl -L -C - -o <filename> <href>`` and
``umbra download ITEM --asset ASSET --dest DIR``. ``bytes`` is the STAC
``file:size`` or ``size`` when that is a non-negative integer; otherwise
the host HEADs that href for ``Content-Length``, with no redirects and
no retries. A href that is not on that bucket is not requested:
``how_to_fetch.curl`` and ``how_to_fetch.umbra`` are null, ``bytes`` is
null, and ``note`` says the commands were omitted. A 404 or 410 from
an allowlisted HEAD raises. Any other HEAD failure, including a full
quick-request slot, leaves ``bytes`` null and still returns the plan.
The hosted server never transfers the file. A local
``uvx --from 'umbra-py[mcp]' umbra-mcp`` server downloads to your own
disk. ``confirm=False`` on this host returns the same size-and-hint
object and does not write a file. It HEADs for ``Content-Length`` only
when the href is https on the public open-data bucket, with no retries,
and it does not follow redirects. A 404 or 410 raises, and so does any
other non-redirect HEAD error. A redirect leaves ``bytes`` null. A href
outside the bucket is not requested; ``bytes`` is the STAC ``file:size``
or ``size`` when that is a non-negative integer, otherwise null. A local
server still HEADs the href and follows redirects.
``url`` is a STAC JSON URL, or a product filename
(``2025-05-17-17-43-46_UMBRA-08_GEC.tif``) or acquisition stem
(``2025-05-17-17-43-46_UMBRA-08``). A filename or stem is resolved through
the catalog index the same way as ``get_item``. The hosted server already
has that index. A local server needs ``umbra index fetch``.
``asset`` selects the file. All of these work:
- a product type, any case: ``GEC``, ``gec``, ``SICD``
- a schema v2 key, including a collision suffix: ``gec``, ``gec_2``,
``csi``, ``sidd``, ``csi_sidd``, ``sicd``, ``cphd``, ``metadata``,
``stac_metadata``, ``di_gif``, ``file``, ``file_tif``
- the original filename (the asset's ``umbra-py:filename``)
``GEC`` and ``gec`` are the primary file. ``gec_2`` is the second GEC,
not the primary. A name that matches nothing raises with the item's v2
keys listed (``gec, sicd, cphd``, …), not the uppercase product types.
Report only the acquisitions **new** since the last check of this site.
SAR's value for monitoring is its cadence, so the natural workflow is
*standing*: run the same search each time and act only on what changed. This
tool packages that idempotent delta — call it in one turn, come back a day
(or a session) later and call it again, and it returns just the passes
published in between, not the whole list again. The scheduler is you: ask the
agent to check the site, and it reports the delta.
Define the site with the same filters as ``search_catalog`` — ``place``
(geocoded to a bbox), ``area`` (task/site-name substring, loosened by
``fuzzy``), ``bbox``, ``products``, and ``start``/``end`` date bounds. The
watch is identified by a stable ``name`` derived from those filters (pass an
explicit ``name`` to run several independent watches over the same site);
``reset=True`` re-establishes the baseline, reporting everything as new.
State persists in the local catalog index's ``meta`` table (created on first
use), so a watch survives across sessions with no extra setup. ``local``
selects the search backend exactly as in ``search_catalog`` — leave it unset
to use the on-disk index when present and a live S3 walk otherwise; for true
monitoring a live walk (``local=false``) catches freshly published passes.
**No model is called** — this is pure set arithmetic over the deterministic
search — and the returned ``new_items`` are context cards ready to hand
straight to ``change_composite`` / ``timescan``, closing the standing-analyst
loop (new pass → composite → describe) without leaving the conversation.
Find archived scenes that **look like** the acquisition at ``url``.
This is the one search that lives in the pixels, not the metadata: where
``search_catalog`` filters by date/bbox/task-name, this renders the query
item's quicklook, embeds it, and ranks the pre-embedded scene index by visual
(cosine) similarity — "find scenes that look like this flooded field / crowded
berth / runway". The query item is excluded from its own results.
Requires a scene-embedding index built ahead of time with ``umbra embed
build`` (a sidecar ``catalog.embed.db`` beside the catalog index) and a
user-supplied multimodal embedding key (``OPENAI_API_KEY``, optionally
``OPENAI_BASE_URL`` / ``UMBRA_SCENE_EMBED_MODEL``) — the only model call is
turning the query image into a vector; ranking is deterministic. ``asset``
picks which product's quicklook to embed (default ``GEC``; match how the index
was built); ``model`` must match the index's embedding model. Returns the
ranked matches as compact records (each with the acquisition's ``href``, so
hand it to ``get_item`` / ``quicklook`` / ``change_composite`` next) plus the
attribution line.
Find archived scenes matching a plain-language ``query`` over their imagery.
Text-to-scene search: embeds ``query`` ("ships at a berth", "a flooded field")
and ranks the stored *image* vectors by cosine similarity. It only works when
the index was built and this query is embedded with the **same** joint
CLIP-family model whose text and image encoders share a vector space (a
dimension mismatch is reported as an error; ``model`` selects it).
Requires the same prebuilt ``catalog.embed.db`` (``umbra embed build``) and
embedding key as ``find_similar``; the only model call is turning the text
query into a vector. Returns the ranked matches as compact records (each with
the acquisition's ``href`` for ``get_item`` / ``quicklook``) plus attribution.
Return a SAR-literate reading of the scene at ``url``.
The picture and the metadata card are produced deterministically; a model
only interprets them. How that model is reached depends on whether this
process holds a vision key:
- **No key** (the public community host): this tool does **not** call a
vision model. It returns the PNG plus a reading kit (SAR rules, the
metadata card, the JSON schema). Produce the JSON, then call
:func:`stamp_description` — that is the validation boundary
(provenance-stamped ``{summary, observed_features, confidence,
caveats}``). Until you stamp it, your prose is not pipeline output.
A baked preview is typically 512 px; the kit names the size and
that a local server can stream a higher-resolution GeoTIFF.
- **Key configured** (local stdio): sends the picture to that vision
model, validates through ``parse_description``, and returns the
stamped object directly.
The packaged SAR rules are the value over just looking at ``quicklook``:
brightness is backscatter not sunlight, speckle is not structure, a dark
patch may be radar shadow not water, one frame is not change.
``preview`` defaults to ``"auto"``: a baked catalog preview if this server
has one, else a fresh render. ``"render"`` is refused on the public host.
Return a grounded reading of *what changed* between passes.
Sibling of :func:`describe_scene`. The composite and the per-block dB grid
are produced deterministically; a model only narrates change those numbers
support. Pass the STAC URLs in time order; mixed polarizations are refused.
This tool **streams Umbra COGs** to build the composite, so a public
community host refuses it (use a local ``uvx --from 'umbra-py[mcp]'
umbra-mcp`` server, or describe each pass with :func:`describe_scene` /
:func:`quicklook` on baked previews). On a local server:
- **No vision key:** returns the composite as an MCP image block plus a
reading kit (SAR change primer, the dB grid, the JSON schema). **You**
(the client's model) narrate, grounded in the numbers.
- **Key configured:** sends picture + grid to that vision model, validates
through ``parse_narration``, and returns a provenance-stamped
``{summary, changes[], confidence, caveats[], change_stats}`` object.
Validate a client-produced scene reading and stamp provenance.
This is the interpretation boundary for the no-key kit path: you read the
picture ``describe_scene`` returned, produce
``{summary, observed_features, confidence, caveats}``, and pass it here.
The library checks the shape, stamps CC-BY and the AI-interpretation
note (the model never sets those), and — when ``image`` is the record
from the kit — appends a deterministic caveat if the picture was a
smaller baked preview. Until this returns, the reading is unvalidated
prose, not pipeline output.
Validate a client-produced change narration and stamp provenance.
Sibling of :func:`stamp_description` for the no-key ``narrate_change``
kit. Pass the JSON you produced plus the ``change_stats`` grid from the
kit so the stamped object carries the numbers the prose was grounded in.
Search, preview, load, and convert Umbra open SAR data.
Umbra publishes 16–25 cm SAR as CC BY 4.0 open data, but no search API —
only a 17+ TB S3 bucket and a static STAC tree. umbra-py is that layer:
search, preview, download, and analysis-ready arrays without the usual 500
lines of glue. A community STAC API (umbra serve) and MCP server sit on
the same host, so pystac-client and Claude can query the archive with
nothing installed.
Status: v0.1.4. Discovery, download, xarray loading,
SICD → geocoded COG, change/timescan composites, chips, a STAC API
(umbra serve, with a community host), and an MCP server all ship. This is
not an InSAR toolbox (phase is not preserved through convert). Not
affiliated with Umbra Lab, Inc.
from umbra_py import CatalogIndex, to_xarray
with CatalogIndex.from_release() as index:
item = next(iter(index.search(area="Centerfield", product_types=["GEC"], limit=1)))
# Stream a downsampled window over HTTP — no multi-GB download. Needs [load].
da = to_xarray(item, max_size=1024, db=True)
print(item.summary())
If the snapshot is missing, the same search against the live bucket is
UmbraCatalog().search(...) / umbra search --area Centerfield.
What you can do
More detail, options, and caveats live in the
docs.
Search by bbox, place name, polygon, or Umbra task (area=).
--local reads the snapshot; omit it to walk S3.
python
from umbra_py import UmbraCatalog
for item in UmbraCatalog().search(area="Centerfield", product_types=["GEC"], limit=5):
print(item.summary())
Preview without downloading the scene: umbra gallery, umbra quicklook <stac-url> --out scene.png --db, umbra view <stac-url> (full-res tiles),
or umbra change --area Centerfield --out change.png.
Load a geocoded GEC into xarray or a GeoTIFF (to_xarray, to_geotiff,
to_stack). Needs [load].
Convert a SICD to a north-up amplitude COG (sicd_to_geocoded_cog,
umbra convert) — phase is discarded. Needs [convert]. Open products
generally have no radiometric metadata, so --calibrate / --noise-model measured refuse rather than invent numbers. See
limitations and the
complex-product handoff.
Chip scenes into georeferenced ML tiles for SR / ATR-style benchmarks from
open Umbra GEC/SICD: umbra chips --area Centerfield --out chips/. See the
ISR training-set cookbook
and Used in research.
Drive it from an agent. Copy-paste recipes for Claude Desktop and Claude
Code: Connect Claude (MCP).
Zero-install remote MCP (no uvx):
bash
# Claude Code
claude mcp add --transport http umbra https://api.umbra-py.space/mcp --scope user
That command is published to the MCP registry
as io.github.reesehammer/umbra-mcp. STAC for pystac-client / QGIS is
https://api.umbra-py.space/ (not /mcp).
docker compose -f deploy/docker-compose.yml up is the one-command self-host.
What the data looks like
Asset
What it is
Use it for
GEC
Geocoded cloud-optimized GeoTIFF
Map-ready imagery. Start here.
CSI
Color sub-aperture GeoTIFF
Quick-look RGB, not a measurement
SIDD
Geocoded detected image (NITF)
Detected imagery in a standard format
SICD
Complex slant-plane image (NITF). Open archive: RGAZIM/PFA.
Phase-preserving downstream. Download; do not convert.
CPHD
Compensated phase history
Custom formation outside umbra-py (download; do not convert). Not an image.
umbra-py downloads SICD/CPHD. umbra convert geocodes a SICD to
amplitude and discards phase. It does not form interferograms or
compute coherence. For a processor that needs the complex pixels, see
Complex products (SICD/CPHD).
Data license & attribution
Umbra's imagery is CC BY 4.0. If you use or redistribute the data or
derived products you must attribute Umbra. The copyright notice Umbra's
terms require is quoted under Citing umbra-py.
umbra-py also stamps this shorter line on its own outputs:
Contains Umbra open data, licensed under CC BY 4.0.
umbra-py itself is Apache 2.0 (LICENSE). The two licenses
are independent and compatible.
Cite all versions with the concept DOI
10.5281/zenodo.23676715
(Zenodo's Cite all versions? value). It resolves to the latest archived
release. The v0.1.4 version DOI is
10.5281/zenodo.23676716.
Machine-readable metadata lives in CITATION.cff
(GitHub's Cite this repository button) and
.zenodo.json (what Zenodo archives).
bibtex
@software{hammerstrom2026umbrapy,
author = {Hammerstrom, Reese},
title = {umbra-py},
year = {2026},
version = {0.1.4},
doi = {10.5281/zenodo.23676715},
url = {https://github.com/reesehammer/umbra-py},
license = {Apache-2.0}
}
Canopy API token. Set it and the same search tools query Umbra's authenticated commercial archive instead of the open bucket; leave it unset for the open data, which needs no credentials.
UMBRA_INDEX_DB
Path to a local catalog index (built with 'umbra index build' or fetched with 'umbra index fetch'). Without one the server walks the public bucket per search, which is correct but slow.
ANTHROPIC_API_KEYsecret
Optional. With a key, describe_scene and narrate_change call a vision model on the server. Without one they return a reading kit (picture + SAR primer) so the client's own model does the reading. A public host must not set this.
OPENAI_API_KEYsecret
Alternative provider for the same two model tools, and the encoder behind find_similar_text.