OSINT Academy — A Subthesis Resource

Open-Source Intelligence — From Curiosity to Capability

A structured curriculum for practitioners who need to find, verify, and document what public information actually shows — without cutting legal or ethical corners.

Start the methodology Browse the tool directory

Start here

Pick a pathway by where you are today. Each one lands you on material pitched at your level.

Beginner

You are new to OSINT

Learn the intelligence cycle, the vocabulary, and the habits that keep an investigation honest before touching a single tool.

Read the methodology →
Intermediate

You work in a specific domain

Journalism, academic research, financial, corporate, or legal — start from the domain guide that matches the work you already do.

Open the domain guides →
Advanced

You want applied examples

Read worked investigations — how a lead became a pivot, how evidence was preserved, and where the work would have failed without discipline.

Study the case studies →

Choose your domain

OSINT is a general discipline with specific expectations in each field. Pick the guide that matches how your findings will be used.

OSINT for Journalism

Source verification, social media investigation, and standards of proof for published reporting.

Financial OSINT

Beneficial ownership, sanctions screening, and following money through public data.

Corporate OSINT

Due diligence, supply-chain investigation, and third-party risk research.

The methodology in four phases

Every serious open-source investigation follows the same arc. Skipping a phase is how investigators end up publishing a fabrication.

  1. Phase 1

    Planning

    Define the question, the audience, and the standard of proof before a single query.

  2. Phase 2

    Collection

    Gather evidence from public sources, documented with timestamps and provenance from the first click.

  3. Phase 3

    Analysis

    Corroborate, triangulate, and stress-test the findings against alternative explanations.

  4. Phase 4

    Reporting

    Communicate what the evidence supports, flag what it does not, and preserve the chain back to source.

Latest tutorials

All tool tutorials →

Featured case studies

All case studies →

Verifying a Viral Image

journalism

  • reverse-image
  • metadata
  • wayback-machine

A viral image presented as recent was matched to an earlier publication, its original context recovered, and the reposter's claim retracted.

Tracing Shell Company Ownership

corporate

  • company-registry
  • whois
  • foia
  • metadata

A five-jurisdiction corporate chain was mapped, with three natural persons identified as common directors across the chain.

Following the Flight Logs

financial

  • metadata
  • foia
  • google-dorking
  • wayback-machine

A published methodology for turning a scanned log set into a searchable, hashed, and cross-referenced dataset suitable for long-form investigation.

From the blog

All articles →