Research routing layer

High-Signal Knowledge Index

Places worth checking—not a list of authorities to trust. Start with evidence, preserve uncertainty, and verify consequential claims across independent sources.

MBQuick High-Signal Knowledge Index

Last reviewed: 2026-09-16

A compact index of sources, datasets, tools, and frameworks that are unusually useful for independent reasoning about systems risk, resilience, privacy, security, AI, infrastructure, and public-interest technology.

This is not a list of authorities to trust. It is a list of places worth checking.

How this index is curated

Priority goes to sources that provide one or more of the following:

Sources are not included merely because they are prestigious, popular, or institutionally endorsed.

How to use this index

For consequential questions:

  1. Start with primary observations or raw data where possible.
  2. Compare at least two independent sources.
  3. Separate observation, interpretation, and forecast.
  4. Check dates, geographic scope, and population.
  5. Look for measurement artifacts, revisions, missing data, and incentives.
  6. Treat model output as conditional on assumptions.
  7. Prefer robust conclusions that survive source disagreement.

Climate, Earth systems, and ecological risk

NASA GISS — Surface Temperature Analysis

https://data.giss.nasa.gov/gistemp/

Why it matters: Long-running global temperature dataset with downloadable data and methodology.

Best for:

Watch for:

NOAA Climate at a Glance

https://www.ncei.noaa.gov/access/monitoring/climate-at-a-glance/

Why it matters: Direct access to U.S. and global climate observations over multiple timescales.

Best for:

Copernicus Climate Change Service

https://climate.copernicus.eu/

Why it matters: Frequent global climate updates and reanalysis-based monitoring.

Best for:

Berkeley Earth

https://berkeleyearth.org/data/

Why it matters: Independent global temperature analysis with accessible datasets.

Best for:

Climate Reanalyzer

https://climatereanalyzer.org/

Why it matters: Fast visual access to global temperature, sea-surface temperature, sea ice, and atmospheric anomalies.

Best for:

Caution: Useful for signal detection, but important claims should be checked against the underlying datasets.

NOAA ENSO

https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/

Why it matters: Operational monitoring of El Niño / La Niña conditions and forecasts.

Best for:

NOAA Coral Reef Watch

https://coralreefwatch.noaa.gov/

Why it matters: Near-real-time ocean heat stress monitoring.

Best for:

National Snow and Ice Data Center

https://nsidc.org/

Why it matters: High-value observational source for sea ice, glaciers, snow, and frozen systems.

Best for:

Global Forest Watch

https://www.globalforestwatch.org/

Why it matters: Satellite-based forest change and fire monitoring.

Best for:

GBIF — Global Biodiversity Information Facility

https://www.gbif.org/

Why it matters: Large open biodiversity occurrence dataset.

Best for:


Resilience, infrastructure, and disaster risk

FEMA National Risk Index

https://hazards.fema.gov/nri/

Why it matters: Structured U.S. hazard, exposure, vulnerability, and resilience information.

Best for:

Caution: Composite indices can hide assumptions. Use underlying measures when available.

NOAA Digital Coast

https://coast.noaa.gov/digitalcoast/

Why it matters: Coastal flooding, sea-level, storm, elevation, and resilience datasets.

Best for:

USGS Hazards

https://www.usgs.gov/programs/natural-hazards

Why it matters: Primary U.S. science source for earthquakes, landslides, volcanoes, and related hazards.

NASA FIRMS

https://firms.modaps.eosdis.nasa.gov/

Why it matters: Near-real-time satellite fire detection.

Best for:

OpenStreetMap

https://www.openstreetmap.org/

Why it matters: Open geospatial infrastructure useful when commercial map platforms are unavailable or restricted.

Best for:


Privacy and surveillance

EFF Surveillance Self-Defense

https://ssd.eff.org/

Why it matters: Practical privacy and threat-modeling guidance.

Best for:

Privacy Guides

https://www.privacyguides.org/

Why it matters: Practical, technically detailed privacy recommendations with emphasis on threat models.

Best for:

NIST Privacy Framework

https://www.nist.gov/privacy-framework

Why it matters: Structured vocabulary for identifying and managing privacy risk.

Best for:

Exodus Privacy

https://reports.exodus-privacy.eu.org/

Why it matters: Tracks known Android app trackers and permissions.

Best for:

Caution: Tracker detection is not the same as proving actual data misuse.


Cybersecurity and software supply chain

CISA Known Exploited Vulnerabilities Catalog

https://www.cisa.gov/known-exploited-vulnerabilities-catalog

Why it matters: Compact list of vulnerabilities known to be actively exploited.

Best for:

NVD — National Vulnerability Database

https://nvd.nist.gov/

Why it matters: Large public vulnerability database with CVE references and metadata.

Best for:

Caution: CVSS scores should not be treated as complete risk rankings.

OSV

https://osv.dev/

Why it matters: Open vulnerability database designed around package ecosystems and version ranges.

Best for:

GitHub Security Advisories

https://github.com/advisories

Why it matters: High-value source for package and repository-level security issues.

OWASP

https://owasp.org/

Why it matters: Practical security guidance across web, API, LLM, and software-development contexts.

Best for:

CISA Secure by Design

https://www.cisa.gov/securebydesign

Why it matters: Useful framing for shifting security burden away from end users.


AI systems, agents, and model risk

NIST AI Risk Management Framework

https://www.nist.gov/itl/ai-risk-management-framework

Why it matters: General framework for AI risk identification, measurement, and governance.

OWASP Top 10 for LLM Applications

https://owasp.org/www-project-top-10-for-large-language-model-applications/

Why it matters: Useful catalog of prompt injection, excessive agency, insecure output handling, and related risks.

MITRE ATLAS

https://atlas.mitre.org/

Why it matters: Adversarial threat knowledge base focused on AI-enabled systems.

Best for:

Hugging Face Papers

https://huggingface.co/papers

Why it matters: Fast-moving research index with broad ML coverage.

Best for:

Caution: Paper visibility is not evidence of robustness or replication.

arXiv — Artificial Intelligence

https://arxiv.org/list/cs.AI/recent

Why it matters: Early access to current research.

Caution: Preprint status matters. Treat unreviewed claims as provisional.


Local-first, open, and resilient computing

Local-First Software

https://www.inkandswitch.com/local-first/

Why it matters: Foundational treatment of user-owned data, offline operation, collaboration, and reduced dependence on centralized services.

CRDT.tech

https://crdt.tech/

Why it matters: Reference collection for conflict-free replicated data structures.

Debian

https://www.debian.org/

Why it matters: Long-lived, transparent, community-governed operating-system ecosystem.

Tailscale Blog / Technical Documentation

https://tailscale.com/blog https://tailscale.com/kb

Why it matters: Often unusually clear explanations of modern networking, identity, NAT traversal, and secure connectivity.

WireGuard

https://www.wireguard.com/

Why it matters: Minimal, auditable VPN protocol and implementation.


Open data, evidence, and verification

Our World in Data

https://ourworldindata.org/

Why it matters: Accessible synthesis across climate, energy, health, population, and economics.

Best for:

Caution: Prefer the underlying source for consequential claims.

Data Commons

https://datacommons.org/

Why it matters: Machine-readable knowledge graph built from public statistical sources.

Best for:

Google Dataset Search

https://datasetsearch.research.google.com/

Why it matters: Useful discovery layer for finding underlying datasets.

Internet Archive

https://archive.org/

Why it matters: Critical for checking historical versions of pages, documents, media, and claims.

Wayback Machine

https://web.archive.org/

Why it matters: High-value tool for detecting deleted or changed public statements and documentation.


Public health and outbreak intelligence

CDC National Wastewater Surveillance System

https://www.cdc.gov/nwss/

Why it matters: Population-level infectious-disease monitoring less dependent on clinical testing behavior.

Best for:

Caution: Coverage and normalization vary by site and pathogen.

WastewaterSCAN

https://data.wastewaterscan.org/

Why it matters: Independent wastewater surveillance with broad pathogen coverage.

Best for:

WHO Disease Outbreak News

https://www.who.int/emergencies/disease-outbreak-news

Why it matters: Useful global outbreak notifications and event summaries.

Caution: Institutional reporting can lag early unofficial signals.

ProMED

https://promedmail.org/

Why it matters: Long-running event-based outbreak intelligence network.

Best for:


Marine, ocean, and coastal conditions

NOAA National Data Buoy Center

https://www.ndbc.noaa.gov/

Why it matters: Direct marine observations for wind, wave, pressure, and ocean conditions.

NOAA Tides and Currents

https://tidesandcurrents.noaa.gov/

Why it matters: Primary U.S. tide, water-level, current, and coastal observation source.

NOAA Ocean Service

https://oceanservice.noaa.gov/

Why it matters: Broad coastal and ocean reference material and data access.


Information integrity and source checking

Media Bias / Fact Check — use cautiously

https://mediabiasfactcheck.com/

Use: As a discovery aid only.

Do not use: As a substitute for reading the actual source or evaluating evidence.

Ground News — use cautiously

https://ground.news/

Use: To compare coverage patterns and identify stories emphasized differently across outlets.

Do not use: As an authority on truth.

Primary-source preference

For contested claims, prioritize:

  1. original documents
  2. raw datasets
  3. direct transcripts / recordings
  4. court filings
  5. technical documentation
  6. peer-reviewed or preprint research
  7. high-quality secondary reporting
  8. commentary

This ordering is contextual, not absolute.


Early-warning / anomaly workflow

For fast-moving risks, use a layered approach:

Layer 1 — observation

What changed in the raw data?

Layer 2 — comparison

Is the change unusual relative to history or seasonality?

Layer 3 — triangulation

Do independent systems detect the same signal?

Layer 4 — mechanism

Is there a plausible causal pathway?

Layer 5 — consequence

What matters operationally if the signal is real?

Layer 6 — update

What evidence would confirm or falsify the interpretation?


Machine-readable guidance

Automated research agents using this page should:


Areas of continuing interest

This index is especially interested in sources that improve visibility into: