Off-Nadir Delta
OfficialOff-Nadir Delta is a live OSINT/GEOINT intelligence server that takes an event from first signal to a collection-ready satellite imagery plan.
Discover events:
query_signalsreturns geolocated, AI-enriched world events (severity, GEOINT score, collection priority) filterable by bbox, date, category, market exposure, observability, and tasking-readiness;query_statsandquery_hotspotsgive aggregate trends and density hotspots.Get situational briefs:
get_world_briefreturns a free AI-synthesized Daily World Brief with developments, why-it-matters, and freshness status.Verify and track reporting:
query_claimsaudits the claim ledger (evidence class, independent source families, downgrades), and event-thread/development tools show what actually changed versus what is merely new.Plan imagery collection:
plan_event_imageryruns a deterministic Sentinel-1 SAR + Sentinel-2 optical plan for one event;search_imageryfinds real pre/post scenes;rank_imaging_prioritytriages where and at what resolution class to collect;survey_observable_eventschecks what a given sensor can actually resolve.Time satellite access:
predict_satellite_passespropagates orbital elements for 7 satellite families (free systematic vs commercial taskable) with off-nadir angles, lighting, and pass windows.Run analysis:
assess_signalgives an AI remote-sensing deep-dive with an actionable imagery handoff;ask_analystruns a durable async agentic brief (poll viaget_analyst_job, charged once on completion).Stand up continuous watching:
create_standing_orderre-answers a question over an area and notifies only on change (free to create, metered per fire);create_monitored_areacontinuously measures metrics (ships, fires, water, vegetation, etc.) from new Sentinel/VIIRS acquisitions, withlist_/get_monitored_areafor history and anomaly flags.Manage credentials and cost:
get_usagepre-flights token balance and plan capabilities; free tools exist for briefs, listings, and status.
This server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
Off-Nadir Delta MCP
From first signal to satellite evidence. Give an AI agent a live intelligence workflow: discover an event, verify the reporting, identify what evidence is missing, determine whether a satellite can actually resolve it, and produce a collection-ready plan.
Real-time event and geospatial intelligence for OSINT, geopolitical risk, and GEOINT work — source-linked, geolocated, and current, not a training snapshot.
23 tools · MCP server version 5.11.0 · full reference
What it does
Seeing the event is easy. Knowing what you can prove — and what evidence is still missing — is the product. Delta:
monitors emerging world events and geolocates them
checks claims against source-linked open data, and reports the source breadth behind each one
assesses whether the event is observable from space, at what resolution, and by which sensor
returns a collection-ready recommendation: what to collect, where, and when
Related MCP server: SkyFi MCP Server
Connect
Hosted server — recommended
No install. Point any MCP client at the remote endpoint and approve once over OAuth 2.1 (Dynamic Client Registration is supported, so most clients need only the URL):
https://offnadir-delta.com/api/v1/mcpIn the Claude or ChatGPT apps: Settings → Connectors → Add custom connector, paste that URL. Cursor and VS Code have one-click installs on the docs page.
Claude Code:
claude mcp add --transport http off-nadir-delta https://offnadir-delta.com/api/v1/mcpThis package — local stdio
Use this when your client cannot do remote OAuth, or you would rather hold an API key in an environment variable and pin a version. Get a key at offnadir-delta.com/account/api.
{
"mcpServers": {
"off-nadir-delta": {
"command": "npx",
"args": ["-y", "offnadir-delta-mcp"],
"env": { "OFFNADIR_DELTA_API_KEY": "ond_..." }
}
}
}Docker:
docker run -i --rm -e OFFNADIR_DELTA_API_KEY=ond_... offnadir-delta-mcpThe package forwards tools/call to the hosted server with your key. tools/list answers
locally from a generated catalog, so registries can introspect it without credentials.
First call — free
12 of the 23 tools cost nothing, so the first thing you run is free:
Give me the latest Daily World Brief. Lead with the three most significant developments, explain why each matters, and cite the supporting signals.
Then try the full signal-to-satellite workflow:
Find the highest-priority observable event from the last 24 hours, verify the reporting, and recommend the best available satellite collection plan.
Tools
Discover
What is happening, what changed, and what today looks like.
Tool | What it does | Cost |
| Geolocated world events from news, hazard feeds and satellite measurements, AI-enriched with severity/GEOINT scores and collection recommendations. | 3–5 tok |
| AI world brief: daily (prior UTC day, all plans) or weekly/monthly by plan. | free |
| What actually CHANGED about the events in an area, not which articles are new. | 3–5 tok |
| One event end to end: state plus every change in order — the "new event or update" distinction a feed cannot make. | free |
| Find a named place (airport, base, plant, port, dam, strait…) in any language or by IATA/ICAO code. | 1 tok |
| What has happened at one place: the registry record plus its most recent linked events, each with HOW it was linked. | 1 tok |
| The relations recorded for one event: others at the same registry facility, naming the same place, or stored as the same campaign, each saying what is shared; plus reports merged into it. | 1 tok |
| Where a warship was reported in the last 30 days (official releases, U.S. | free |
Plan
Whether a satellite can resolve it, which one, and when it next passes.
Tool | What it does | Cost |
| Search the imagery catalog over an area and window. | 2 tok |
| The deterministic imagery plan for ONE event: checks sentinel-1-grd (SAR, the only look that survives cloud and night) and sentinel-2-l2a exactly once each — plus NISAR L-band for flood_mapping (water under tall vegetation) — against the event footprint and a pre/post window. | 4 tok |
| WHERE, and with what class of satellite, observation is most worthwhile now: composite IMPORTANCE crossed with the SPEC CLASS the required resolution demands — coarse, hr (free Sentinel-class) or vhr. | 1 tok |
| WHEN a place can next be imaged and by WHAT: 13 free-systematic and commercial-taskable families. | 2 tok |
Analyze
Turn reporting into a cited assessment you can audit afterwards.
Tool | What it does | Cost |
| AI remote-sensing deep-dive for one signal: what to observe, sensors, a collection window. | 5/15 tok |
| Ask the Delta Analyst an OSINT/GEOINT question; returns a structured brief. | 5–123 tok |
| Status and result of an ask_analyst run. | free |
Watch
Stand up continuous coverage and be told only when the answer changes. Creating and listing cost nothing — you are metered only when a check actually fires, or a new acquisition is actually measured.
Tool | What it does | Cost |
| The Watchlist as one list, each entry with a state bucket. | free |
| What changed across the Watchlist, newest first: one row per change a watch reported, with read state. | free |
| One watch end to end: target, state, latest change, measurements with recent series, standing-order questions, and for an event watch its stage, developments, imagery check and thread, plus this account's notes. | free |
| Add a target to the Watchlist. | free |
| Pause, resume or close a watch. | free |
| Delete a watch and its underlying resources — bound monitored areas with their history, and standing orders. | free |
| Add a note to a watch, start a thread, or reply. | free |
Account
Pre-flight your token balance before a metered call.
Tool | What it does | Cost |
| The calling key's remaining token balance and plan capabilities. | free |
Pricing and metering
Usage-based Delta tokens, identical over REST and MCP. Costs above are generated from the same constant the biller reads, and every metered result carries the exact charge it incurred — so you never have to trust a number in a README. Free tools never touch the balance.
ask_analyst is hybrid-async: it enqueues a run and returns a job_id immediately; poll
get_analyst_job (free — the run is charged once, on completion). Most questions finish in
~30–90s; a brief with satellite-imagery lookups or many sources can take 2–3 minutes.
API and MCP access. Available on every plan, including Free. Calls are metered on your token balance, and what each call returns follows your plan — a call your plan does not open says which plan does. Call get_usage (free) to pre-flight. Current limits and package prices:
offnadir-delta.com/pricing.
Authentication
Two credentials are accepted on the hosted endpoint:
OAuth 2.1 with PKCE and Dynamic Client Registration — for interactive clients
Bearer API key (
Authorization: Bearer ond_...) — for servers, CI, and this stdio package
Keys are shown once at creation, hashed at rest, and revocable at any time from /account/api.
Resources
Resource | Description |
| The most recent AI-synthesized Daily World Brief (JSON). Free. |
| JSON Schema of the public Signal shape returned by query_signals / /api/v1/signals. |
| Remaining token balance and plan capabilities for the calling key. Free. |
| How current the data is (ingestion/enrichment frontier), the Daily World Brief status, and an Operational/Delayed/Degraded roll-up. Free. |
| The Daily World Brief for a specific UTC date (YYYY-MM-DD). Free. |
| A Watchlist entry with its current state, latest meaningful change, measurements, (for event watches) the full event thread, and the notes kept against it. The same body get_watch returns. Free. |
Links
Live server version probe (no auth, no tokens)
License
Apache-2.0
Available Tools
20 toolsask_analystADestructiveInspect
Ask the Delta Analyst an OSINT/GEOINT question. Runs an agentic multi-step analysis over the signal corpus and returns a structured brief (summary, findings with collection recommendations, assessment, citations). Costs 5–123 tokens (usage-based, metered by the compute the question actually uses; charged ONCE, when the run completes; the exact charge and remaining balance are in the result meta.tokens). Durable async: the run is ENQUEUED and returns {status:"processing", job_id} immediately, then completes in a background worker — so it is never lost to a client timeout. Timing: most questions finish in ~30–90s; a complex brief (satellite-imagery lookups or many sources) can take 2–3 minutes. Fetch the finished brief by calling get_analyst_job with the job_id (poll every ~10–20s), or ask_analyst again with the SAME idempotencyKey (no second charge).
Inputs: question (required), bbox, mode, idempotencyKey, +1 more. Returns: meta, brief, job_id, status, +3 more
assess_signalADestructiveIdempotentInspect
Run an AI RS (remote-sensing) deep-dive assessment for a specific signal: what to observe, recommended sensors, and a collection window. eventId is the id from query_signals. The result also carries a deterministic context block (event id/date, normalized target, AOI bbox, observability + quality verdict, and an imagery_handoff giving the exact bbox + event_date to pass to search_imagery for REAL pre/post scene candidates) — turning the assessment into an actionable collection plan. Costs 5 (quick) or 15 (deep) tokens, charged to the key owner's balance. A prior assessment for the same signal is cached (no re-charge). The exact charge and remaining balance are in the result meta.tokens. Signals that are not satellite-observable (observability:"not-observable" — e.g. political statements or broad-area events with no imageable physical mark) are rejected BEFORE any charge, so pre-filter with query_signals observability:"observable" to spend only where imagery helps.
Inputs: eventId (required), kind. Returns: kind, meta, model, cached, +1 more
create_monitored_areaAInspect
Put a place under continuous satellite measurement: pick an area and what to count, and every new Sentinel-1 / Sentinel-2 / VIIRS acquisition over it is measured automatically from then on. Use this when the question is about a quantity at a fixed place over time ("how many ships are alongside", "how much has burned", "is the water receding") rather than about events, which is create_standing_order. Creating is free; each automatic check costs a small number of tokens only when it finds new imagery.
Inputs: bbox (required), metric (required), name, start_date. Returns: area, summary, metering
create_standing_orderAInspect
Put an area under CONTINUOUS watch: save a question plus a bounding box and Delta re-answers it on a schedule, notifying only when the answer actually changed. Creating one is FREE. Each time it fires it runs ask_analyst and is metered like any Analyst question, so the cost is per CHANGE, not per check: a deterministic pass over the corpus decides whether anything new crossed the reporting bar, and quiet periods never invoke the model or charge anything. Returns projected_monthly_tokens_max — the ceiling if every single check fired — so the cost is visible before committing. Cadence and how many orders you may hold are set by your plan; the error says which limit you hit. Use it when the question is "tell me when this changes" rather than "what is happening right now".
Inputs: bbox (required), name, cadence, question, +4 more. Returns: order, summary, projected_monthly_tokens_max
delete_standing_orderADestructiveIdempotentInspect
Delete a standing order by id, or pause/resume it instead by passing active=false/true. Pausing keeps the order and its history; deleting removes both. Neither costs tokens. A paused order still counts against the plan limit, so delete rather than pause when you want the slot back.
Inputs: order_id (required), active. Returns: order, deleted, summary
get_analyst_jobARead-onlyInspect
Fetch the status and result of an ask_analyst run by job_id. Job status is ONLY "processing" (still running — poll again in ~10-20s), "done" (with the finished brief and meta.tokens), or "error". When done, result_quality.status ("complete" | "partial", with any issues[]) reports whether the structured output is fully populated — this is SEPARATE from the job status (a done job can carry a partial result). Free of token charges — the run itself is charged once on completion. A job is visible only to the API key owner that created it.
Inputs: job_id (required), response_format. Returns: meta, brief, error, job_id, +5 more
get_monitored_areaARead-onlyInspect
Fetch one monitored area with its full measurement history — every acquisition that was measured, its value, and whether it was flagged anomalous. This is the time series behind the number that list_monitored_areas reports, so use it to answer "is it going up", "when did it change", or "how unusual is today". Anomaly flags come from a median-absolute-deviation test on the series, not a fixed threshold. Free of token charges.
Inputs: area_id (required). Returns: area, summary, metering
get_usageARead-onlyInspect
Check the calling key's remaining token balance and plan capabilities — monthly allocation, tokens used this period, tokens remaining, and whether the plan includes AI tools over the API (assess_signal / ask_analyst). Use this to pre-flight a metered call: decide whether enough balance is left before spending. Free of token charges.
Inputs: no named parameters. Returns: plan, tokens
get_world_briefARead-onlyInspect
Fetch the Daily World Brief — an AI-synthesized OSINT/GEOINT digest of the previous UTC day's worldwide event signals (headline, executive summary, top developments with why-it-matters and what-to-watch, per-theme roll-up, ranked signals). Free of token charges. The result includes a freshness object (brief_date, age_hours, is_stale) — if is_stale is true this is the latest published brief and a newer day is not yet available, so relay it as possibly out of date.
Inputs: date. Returns: brief, freshness
list_monitored_areasARead-onlyInspect
List the places under continuous satellite measurement on this key (Delta Monitor), with each area’s metric, most recent value, change since the previous measurement, whether that value was flagged anomalous, and coverage — how many acquisitions were measured versus how many exist. Coverage window_total is null when the catalog total is UNKNOWN; null never means zero. Also returns how many areas the plan allows and how many remain. Free of token charges.
Inputs: no named parameters. Returns: areas, limits, summary, metering
list_standing_ordersARead-onlyInspect
List the standing orders on this key, with each one’s cadence, watched area, when it last checked, when it last actually fired, and how many consecutive checks found nothing (quiet_checks — a high number means the watch is not earning its place). Also returns how many orders the plan allows and how many remain. Free of token charges.
Inputs: no named parameters. Returns: limits, orders, summary
plan_event_imageryADestructiveInspect
Plan the imagery evidence for ONE event in a single call, instead of guessing collections one at a time. Give the event_id and what you are trying to establish (damage_assessment | flood_mapping | wildfire_assessment) and the SERVER runs the deterministic plan for that goal: it always checks BOTH in-app sensors — sentinel-1-grd (SAR: all-weather, the only look that survives cloud and night) and sentinel-2-l2a (optical, human-legible) — exactly once each, against the event’s own footprint and a pre/post window around its date. The result carries, per search, why it was made, how many scenes came back, how many actually COVER the event and are usable (cloud-obscured optical does not count), the SAR pair status, and whether a pre/post bracket exists. It also states screening / detection / identification capability: Sentinel-1/2 screen and detect at facility scale, and only object-level identification needs commercial VHR — a VHR recommendation NEVER invalidates what the free catalog already showed. Prefer this over several search_imagery calls for the same event: it cannot miss the SAR look and cannot repeat a search. An event with no resolvable footprint is refused rather than planned against a guess. Costs 4 token(s) per call (it deliberately issues two catalog searches).
Inputs: event_id (required), analysis_goal (required), event_date. Returns: meta, plan, summary
predict_satellite_passesADestructiveInspect
WHEN can this place next be imaged, and by WHAT — the timing half of collection planning. Propagates current orbital elements (SGP4 over day-cached CelesTrak two-line elements) for seven families and returns the access windows over a target: Sentinel-1 (SAR), Sentinel-2 and Landsat (free, SYSTEMATIC — routinely collected, so near-certain), plus WorldView, ICEYE, Capella and SkySat (commercial, AGILE — taskable ACCESS opportunities that require a paid order and are NOT guaranteed collects). Each pass carries acquisition/loss times, the closest-approach instant, peak elevation, OFF-NADIR angle, ground distance, ascending/descending, solar elevation and whether the target is sunlit (optical needs light; SAR does not), and the age of the element set it was computed from. Use it to answer "when is the next chance to see this", to choose between waiting for a free systematic pass and paying to task an agile one, and to time a pre/post change-detection pair. Give lat/lon, or a bbox whose centre is used. Horizon is capped at 7 days. If elements cannot be retrieved the result says so (retrieval_ok:false) — that means timing is UNAVAILABLE, never "no passes". Every pass is a GEOMETRIC access opportunity computed from orbital elements and swath width (geometry_only:true, acquisition_plan_verified:false): no operator collection plan is consulted, so this is when a sensor COULD see the target, never a confirmed acquisition schedule. With no start_date the window begins NOW, so the first pass listed is always still ahead. Costs 2 token(s) per call.
Inputs: lat, lon, bbox, end_date, +3 more. Returns: meta, passes, summary, freshness
query_claimsARead-onlyInspect
Read the LEDGER of claims this key has been given — every factual assertion the Analyst made, with its evidence class (CONFIRMED / REPORTED / PARTY_CLAIM / ASSESSMENT), how many INDEPENDENT source families backed it, and the publishers. The point is the time axis: when a later answer restated the same assertion, the claim carries the link and says which way the evidence moved — restated_only surfaces those chains, and downgraded_only isolates the cases where a later answer was LESS sure than an earlier one, which is where this product contradicted itself. Use it to audit what you were told before acting on it, or to check whether an assertion has since weakened. Scoped to your own key; no other caller’s claims are visible. Free of token charges.
Inputs: limit, since, restated_only, evidence_class, +1 more. Returns: claims, counts, summary
query_hotspotsADestructiveInspect
Geographic hotspots — signal density grid-binned into cells, ranked by event count, each with peak severity, the categories present, and up to 5 representative event_ids (trace a cell back to its signals). Use to find WHERE activity is concentrating. NOTE the unit: cells count satellite-observable points (meta.population = rs_observable_points); meta reports source_point_count and dropped_by_geo_count / dropped_by_severity_count so point_count is fully accountable. Costs 1 token(s) per call.
Inputs: bbox, date, days, limit, +3 more. Returns: meta, summary, hotspots
query_signalsBDestructiveInspect
Query geolocated world event signals (Delta Signals: geopolitical, security, disaster, and infrastructure events distilled from global news media, AI-enriched with severity/GEOINT scores and satellite-collection recommendations). Filter by bounding box, date window, and category. Costs 3 token(s) per call, charged to the API key owner's balance.
Inputs: bbox, date, days, sort, +15 more. Returns: meta, signals, summary
query_statsADestructiveInspect
Aggregate statistics over the signal corpus — total event count plus per-category and per-day breakdown (trend) for a bounding box and date window. Cheaper than query_signals (returns roll-ups, not rows). NOTE the unit: total counts article-deduped events (meta.population = article_deduped_events), which is NOT cluster-collapsed, so it is >= the query_signals count for the same window. Costs 1 token(s) per call.
Inputs: bbox, date, days, categories. Returns: meta, stats, summary
rank_imaging_priorityADestructiveInspect
WHERE — and with what class (and therefore cost) of satellite — is observation most worthwhile right now? Crosses each event’s composite IMPORTANCE (severity × source breadth × market relevance) with the SPEC CLASS its required resolution demands: coarse (≤100 m), hr (≤10 m, Sentinel-class free data) or vhr (sub-metre, commercial tasking). Returns how many imageable events fall in each class with their mean importance, plus the top targets with importance, required class and AOI. Use it to triage a theatre before spending on imagery: a high-importance event that only needs hr is answerable with free Sentinel data, while a vhr one is what a paid order is for. Deterministic — no LLM, no per-event cost. Costs 1 token(s) per call.
Inputs: bbox, top_n, end_date, categories, +2 more. Returns: meta, summary, priority
search_imageryADestructiveInspect
Search the satellite imagery catalog (Sentinel-1/2, OPERA RTC-S1) for scenes over an area and date window — the natural follow-up to a signal (find imagery over the event location). Returns scene metadata (id, datetime, footprint, cloud cover, platform, orbit geometry, coverage, catalog link) — no imagery bytes. Pass eventDate to classify each scene timing=pre/post/same_day_unknown (a same-day scene is same_day_unknown, never post, without a real event time) and get pre/post bracketing + window_status + SAR sar_pair_status in meta. Pass eventPoint [lon,lat] and/or eventAoi [minLon,minLat,maxLon,maxLat] to get each scene's target_relation (covers_event_geometry = pure geometry gate; usable_for_analysis additionally requires acceptable cloud for optical, so a 99%-cloud scene is geometry-covering but not analysis-usable) — so a scene that only clips the wide bbox is not mistaken for covering the event. Costs 2 token(s) per call.
Inputs: bbox (required), date, days, limit, +7 more. Returns: meta, scenes, summary
survey_observable_eventsADestructiveInspect
Which events in a window can a given in-app sensor actually RESOLVE? Evaluates the FULL set (not just the top few) against each event’s precomputed required resolution, and returns how many are observable vs not, the breakdown by required resolution, and the top observable events with the ready-made imaging rationale. Observability here is resolvability — whether the physical mark is large enough for the sensor: Sentinel-2 is ~10 m optical (needs daylight and clear sky), Sentinel-1 is SAR (all-weather, day or night). Prefer this over asking about events one at a time: it is exhaustive AND cheap, and it is the honest way to answer "what can we actually see" before committing collection effort. The population is deliberately UNGATED by tasking readiness — it answers "what could this sensor resolve", not "what may we task" — so its total sits above rank_imaging_priority and counts a different unit from query_signals' clusters; see population_detail, and read collection_ready per event for taskability. Deterministic — no LLM. Costs 1 token(s) per call.
Inputs: bbox, top_n, sensor, end_date, +3 more. Returns: meta, survey, summary
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 20 tools
Most tools have clearly distinct purposes across signal querying, imagery planning, analyst Q&A, and monitoring. A few adjacent tools (e.g., plan_event_imagery vs search_imagery, rank_imaging_priority vs survey_observable_events) require careful reading of descriptions, but the boundaries are explicit enough to avoid gross misselection.
All tool names follow a consistent verb_noun snake_case pattern (get_, query_, create_, delete_, list_, plan_, rank_, survey_, predict_, assess_, ask_), with no camelCase or mixed conventions.
At 20 tools, the server is broad in scope but the count is on the heavy side. Each tool serves a distinct function, but the overall surface feels dense compared to a typical well-scoped server (3-15 tools). No redundancy, but the large number increases cognitive load.
The server covers core workflows well, but lifecycle coverage is incomplete: monitored areas have create/list/get but no delete or update, and standing orders lack a true update (only pause/resume via delete_standing_order). This creates dead ends for users wanting to modify or remove monitoring. Signal querying and imagery planning are thorough.
Maintenance
Related MCP Connectors
Real-time planetary signal engine and Model Context Protocol (MCP) server for autonomous AI agents.
Hosted MCP server for live public-data APIs and Skills for AI agents.
Geospatial AI MCP server — satellite imagery, embeddings, weather, GNS governance
MCP server for OnceAsk, the AI-native current-address layer for people and agents.
Related MCP Servers
- AlicenseCqualityDmaintenanceMCP server for the Geopera geospatial data platform that enables AI agents to discover imagery, place and manage orders, and run analytics using the same API as other Geopera clients.100MIT
- AlicenseNot gradedqualityDmaintenanceProduction-grade MCP server wrapping SkyFi's satellite imagery platform API. Enables AI agents to conversationally search, price, order, and monitor satellite imagery through the Model Context Protocol.MIT
- FlicenseBqualityDmaintenanceA Python MCP server for discovering and analyzing Sentinel-2 L2A imagery, enabling AI agents to search scenes, check quality, and compute spectral indices like NDVI, NDWI, and NBR from remote COG windows.4-
- AlicenseNot gradedqualityBmaintenanceMCP server for real-time global intelligence, AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking, enabling agents and scripts to access curated news feeds, country instability indices, and market data.90,631AGPL 3.0