Platform Chain of Custody Validation Deployment Company Contact Request a Demo
GhostKey Development Inc. /  Digital Forensics

Track everything.Prove all of it.

Bloodhound is a self-hosted digital forensics platform that turns device and disk evidence into defensible findings, with tamper-evident chain of custody built into every action.

Self-hosted Offline by default Append-only custody ledger Validated parsers
Bloodhound
E01 / EWF•EnCase images RAW / DD•Bit-for-bit AFF4•Advanced forensic format VMDK / VHDX / VDI•Virtual disks iOS BACKUP•Encrypted & plain ANDROID FS•Filesystem & logical LOGICAL COLLECTIONS•Targeted sets E01 / EWF•EnCase images RAW / DD•Bit-for-bit AFF4•Advanced forensic format VMDK / VHDX / VDI•Virtual disks iOS BACKUP•Encrypted & plain ANDROID FS•Filesystem & logical LOGICAL COLLECTIONS•Targeted sets
Custody is the foundation

Most forensic suites treat custody as paperwork

Bolted on after the fact, filled in from memory. Bloodhound treats it as the foundation. Every evidence item is hashed at acquisition and verified on every access. Every custody event, who touched what, when, and why, is written to an append-only ledger where each entry is cryptographically chained to the one before it.

Alter a record and the chain breaks visibly. When your methodology gets challenged, you have a signed, verifiable trail from acquisition to final report.

How the ledger works
custody-ledger — case 2026-0441 CHAIN VERIFIED
# entry 0117 — append-only, signed seq : 117 timestamp : 2026-09-14T22:41:08.331Z actor : j.mercer <examiner, badge 4417> action : EVIDENCE_ACCESS (read-only mount) item : EV-003 / iPhone 14 Pro / full fs reason : "warrant 26-CR-1180 scope review" sha256 : 9f2c41ab…d7e08b31 [MATCH] prev_hash : 1c5d77e4…b30a9e12 entry_hash : e30af95c…8b21d6ff signature : VALID (ed25519 / gk-node-01) $ bloodhound ledger verify --case 2026-0441 ✓ 117 entries walked, 0 breaks, 0 gaps ✓ 12 evidence items, 12 hash matches ✓ chain root anchored 2026-09-02T14:10:55Z
0%
Custody events written to the ledger
0+
Artifact parsers across mobile and desktop
0
Bytes of case data leaving your network
SHA-0
Hashing at intake and on every access
Capabilities

Find what matters

Parsers, timelining and link analysis that get an examiner to the relevant records quickly, and keep the provenance of every one of them.

Forensic image ingest

Open and standard formats, opened read-only from the first byte. Originals are never modified, and analysis runs against a verified working copy.

  • E01/EWF, raw dd, AFF4, split images
  • VMDK, VHD/VHDX, VDI virtual disks
  • iOS backups, Android filesystem and logical

Purpose-built parsers

Messages, call logs, contacts, browser history, location data, media metadata and app databases, including recovery of deleted records.

  • SQLite WAL, journal and freelist recovery
  • Plist, protobuf, Realm and LevelDB stores
  • Carving for unallocated and slack space

Unified supertimeline

One chronology across every artifact and every device in the case, normalised to a single timezone. Each row carries its source.

  • Cross-device event correlation
  • Filesystem, app, system and network events
  • Filter, pivot and bookmark to findings

Link analysis

Maps the relationships between people, numbers, accounts and devices, so the contact that ties two suspects together across three handsets is visible rather than inferred.

  • Entity resolution across identifiers
  • Communication frequency and direction
  • Shortest-path and cluster views

Hash-set matching

Flags known material and filters out known-good files, which is usually the difference between a two-day review and a two-week one.

  • NSRL and custom hash set import
  • Known-bad alerting with categories
  • Fuzzy hashing for near-duplicates

Court-ready reporting

A report builder that assembles findings and custody records into output you can disclose.

  • Findings, exhibits and custody appendix
  • Hash manifest for every attachment
  • PDF, HTML and structured export

The workflow

From seized device to signed report

STEP 01

Acquire

Register the item, capture the acquisition metadata, hash the source before anything else touches it.

STEP 02

Verify

The working copy is hash-matched against the original. A mismatch halts the case.

STEP 03

Parse

Parsers run across the image, extracting and recovering records with their source offsets retained.

STEP 04

Correlate

Timeline, link analysis and hash matching pull the picture together across all items.

STEP 05

Report

Findings and the custody ledger are assembled, signed and exported.

case dashboard — STATE v. R. HOLT
CASE 2026-0441 opened 02 Sep 2026 lead: j.mercer EVIDENCE 12 items 12 verified 0 quarantined CUSTODY 117 events chain intact 0 anomalies PARSERS 41 of 41 complete [OK] ARTIFACTS RECOVERED messages .......... 48,112 3,204 deleted call logs .......... 6,977 812 deleted locations ..........129,440 media .............. 21,338 17 hash-set hits browser history .... 9,106 TIMELINE 214,073 events 02 Jan – 14 Sep 2026 BOOKMARKS 38 findings flagged for report
The workspace

Case status, custody integrity and key findings in one view

The dashboard is the front page of a case. Behind it sit a fast evidence explorer, communications and media review, geolocation mapping and the report builder.

  • Evidence explorer with filesystem navigation and preview
  • Threaded communications review across SMS, RCS and app messaging
  • Media review with EXIF, device attribution and duplicate collapsing
  • Geolocation mapping with track reconstruction and dwell detection
  • Case-wide search with regex, keyword lists and hit provenance
See the full platform

Where Bloodhound sits

A serious alternative for teams whose evidence gets scrutinised

Built for investigators, DFIR teams and legal engagements where the integrity of the evidence is as important as the findings themselves.

Capability Bloodhound Commercial mobile suites Open-source toolkits
Cryptographically chained custody ledgerBuilt inAudit logManual
Hash verification on every accessEvery accessAt intakeAt intake
Self-hosted, air-gap capableDefaultVaries / cloud tiersYes
Mobile extraction parsingFullFullLimited
Unified cross-device supertimelineYesPer-deviceAdd-on
Signed tool validation reportShippedOn requestCommunity
Per-seat cloud dependencyNoneCommonNone
Report includes full custody appendixAutomaticAssembled manuallyManual

Based on our own assessment of typical product tiers, including Cellebrite, Magnet and Autopsy. Feature sets move. Run a side-by-side against your own reference data before you take our word for any of it.

Prove the tool itself works

An autonomous validation suite, shipped in the box

Bloodhound tests its parsers against known reference datasets and produces a signed validation report, giving you documented, defensible proof of tool reliability before anyone asks for it.

validation-run 2026.09.1
$ bloodhound validate --suite all --sign PASS ios.messages.sms 142/142 records PASS ios.locations.cache 88/88 records PASS android.calllog.deleted 64/64 recovered PASS ewf.reader.integrity hash match PASS timeline.normalisation tz correct report: validation-2026.09.1.pdf [SIGNED]

Every artifact accounted for.
Every action on the record.

Bring Bloodhound into your lab on your own hardware. Evaluation licences include the full platform, the validation suite and the reference datasets it runs against.