National Cyber Warfare Foundation (NCWF)login

phantomtide for ranking maritime and airspace anomalies from open signals


0 user ratings
2026-11-16 23:29:24
milo
Red Team (CNA)
"phantomtide

phantomtide is a geospatial OSINT platform that fuses AIS, ADS-B, thermal, and environmental feeds into scored hotspots so authorized analysts can triage what deserves attention, for sanctions monitoring, dark-vessel research, and defensive intelligence work.








Tooltg12/phantomtide — cross-domain maritime and airspace intelligence platform built on open signals, documented via a public docs repo with a hosted product
CategoryGeospatial OSINT / multi-source intelligence fusion
Primary UseTriaging cross-source hotspots across AIS vessel traffic, ADS-B aircraft, DSC radio traffic, thermal detections, and sanctions watchlists in a single ranked workspace
Safe UseIntended for authorized analysts: maritime domain awareness, sanctions-compliance research, search-and-rescue support, and educational OSINT evaluation against publicly available open data
Telemetry NoteThe tool is read-only against open feeds; defenders observing it see only ordinary HTTPS polling of public AIS/ADS-B/EONET sources, plus calls to its documented public restricted-airspace-crossings endpoint

phantomtide attacks a problem every multi-source intelligence workflow eventually hits: when you overlay a dozen open feeds on a map, everything looks equally important, and nothing gets triaged. The project's README frames the entire product around one analyst question — what deserves attention right now, and why — and every design decision it documents serves that question. Rather than rendering raw AIS positions, ADS-B tracks, thermal detections, and advisory feeds as parallel layers of equal weight, the platform scores overlap between sources and ranks convergence zones with explicit contributor weights and evidence counts. That ranking-first philosophy is what separates it from a generic dashboard product.


The first thing any operator should understand about tg12/phantomtide is that the public GitHub repository is deliberately not the application codebase. The README is explicit: the repo hosts public documentation, release notes, screenshots, demo videos, and feedback intake, while the working product runs as a hosted service with a documented release line, currently v2.2.0. That makes this repo an interesting specimen in its own right — a commercial-ish intelligence product that treats its public documentation as the open artifact. For an evaluator, it means the README is unusually dense with behavioral detail, because the docs are the product surface you can actually audit before committing to the hosted workflow.


The scoring model is the analytical core. Scored convergence zones mark geographic areas where multiple independent signals coincide — for example, a thermal detection cluster overlapping with a vessel-coverage gap, or an aircraft track crossing into a restricted airspace geometry. Each zone carries contributor weights and evidence counts, so an analyst can see not just that something is flagged, but which sources drove the flag and how strong each contribution was. This is the correct architecture for triage: it converts a flood of raw open data into a ranked queue, and it makes the ranking auditable rather than presenting a black-box heat blob.


A distinctive architectural choice is the freshness semantics. The README documents six explicit workspace states — Live, Degraded, Stale, Tier-limited, New data available, and Live paused — and insists that degraded or cached data is never flattened into a healthy-looking map. This is rare honesty in a monitoring product. Most dashboards quietly render stale cached feeds as if they were current, which produces confident wrong answers; phantomtide instead keeps the truth of each source's ingest state visible at all times, and the browser defaults to a stable manual workspace that only applies new state when the analyst refreshes or explicitly enables live mode. Live mode even pauses while the analyst is inspecting detail panels or manipulating the map, so the workspace never redraws underneath an active investigation.


On the maritime side, the platform ingests and classifies DSC (Digital Selective Calling) communications into test, safety, routine, distress, and SAR-linked semantic classes — a practical touch that prevents routine radio checks from polluting an incident feed. Distress positions render with a pulsing ring so active distress contacts stay visually distinct at every zoom level, and the distress banner polls continuously in critical-severity color even when the analyst is in map-only view. Vessel intelligence rows also carry dark-vessel, sanctioned, military, and emergency context where available, which is where the tool earns its sanctions and osint topic tags: it is built for the compliance and maritime-domain-awareness analyst as much as the hobbyist tracker.


Aircraft are treated as a first-class analyst workflow rather than a passive ADS-B layer. Tracked aircraft come with mission cues, watchlist context, alert banners, free-text quick jump, and map-focus jumps. The v2.2.0 release adds live GNSS interference cells derived from ADS-B data — a genuinely useful open-signals technique, since GPS jamming and spoofing in a region betray themselves through anomalies in aircraft-reported position quality. That sits alongside a second earthquake catalogue (EMSC) and a NASA EONET natural-hazards layer, extending the platform from purely man-made signals into environmental context that explains why traffic patterns shift.


The thermal-vessel gap layer deserves specific attention because it is the tool's sharpest dark-vessel angle. It scores grid cells where thermal detections and vessel coverage diverge — areas where something is radiating heat on the water but no AIS-reporting vessel corresponds. Each cell exposes gap type, detection count, and vessel count, refreshed every six hours as a premium layer. Layered against the infrastructure map — chokepoints, ports, pipelines, refineries, terminals — this gives an analyst a defensible starting point for dark-vessel investigations without requiring commercial satellite tasking. The Area Intelligence Report then compresses spatial context around any point into a structured briefing artifact, and a proximity query supports fast pivots from anomaly to neighborhood.


Analyst evidence handling matured in v2.0 with the Analyst Notebook: a per-user, session-persistent, private dossier where any visible event — vessel contact, aircraft track, thermal detection, airspace notice, piracy incident — can be saved with a timestamp, free-text note, and user tags. The API surface is documented plainly: GET /api/notebook/entries returns the filterable dossier, GET /api/notebook/export streams a CSV of event_id, source_id, display_name, user_note, and tags, and GET /api/notebook/summary feeds per-source counts to the sidebar. Duplicate saves return the existing entry rather than creating noise — a small detail that says a lot about how the workflow was designed by people who have actually maintained an evidence log. Storage is tier-gated at 50 entries for Starter, 500 for Premium, and unlimited at Enterprise.


The one public machine-consumable endpoint is a model of honest scoping. GET /api/public/aircraft/restricted-airspace-crossings serves replay-derived restricted-airspace crossing candidates, pollable by an sample_after watermark with a deliberately simple response shape of quality, when, who, what, and where. The README is unusually careful about what this feed is not: not a live enforcement alert, not proof of wrongdoing, and not a general archive API. That framing matters for responsible use — crossing candidates are investigative leads for authorized analysts, not accusations. The endpoint can be exercised locally via docker compose up --build against http://localhost, which is the closest thing the repo offers to a runnable artifact, and contract diagnostics are available through ?include_meta=true for freshness and reference-state checks.


The extended coverage set reads like a maritime situational-awareness wish list: tide stations with prediction deviation at strategic chokepoints, low-elevation coastal zone context, live flow across ten GB subsea power interconnectors, LNG terminal send-out with national gas-storage fill, dual-polar daily sea-ice edges, Baltic icebreaker restrictions, and public KiwiSDR HF receivers with TDoA-capable sites marked. Fleet track history renders per-aircraft polylines for a given airline prefix over a rolling window. The unifying pattern is energy-and-chokepoint awareness: the platform is clearly aimed at analysts who need to reason about shipping, infrastructure, and environmental stress as one system rather than as disconnected feeds.


For the authorized professional, phantomtide occupies a specific niche: it is an open-signals triage and briefing layer, not an exploitation tool and not a substitute for commercial satellite or tasking capabilities. Its value is in ranking, context, and honesty about data quality — the three things that most often break open-source intelligence workflows. Watch the tier gating on notebook capacity, fleet tracks, thermal gaps, and ice layers when evaluating against your requirements, and treat the docs repo as the contract: the release cadence (v2.2.1 tracked with DSC geometry registries and analyst filters next) is visible in the README, so you can judge maintenance velocity directly rather than inferring it from commit history that does not exist here.



Official project repository for tg12/phantomtide.

Download Tool

Educational analysis for authorized security professionals. Use only in controlled, authorized environments.






Source: OffensiveSec
Source Link: https://www.offsecblog.com/2026/11/phantomtide-for-ranking-maritime-and.html


Comments
new comment
Nobody has commented yet. Will you be the first?
 
Forum
Red Team (CNA)



Copyright 2012 through 2026 - National Cyber Warfare Foundation - All rights reserved worldwide.