Skip to content

PerfektBlue v2.0.0 — Bluetooth-first adaptive framework

Choose a tag to compare

View all tags
@github-actions github-actions released this 22 Jun 20:43
· 1 commit to main since this release
v2.0.0
7bbafc0

PerfektBlue 2.0

Bluetooth-first. Evidence-driven. Built for real research workflows.

PerfektBlue 2.0 replaces the original proof of concept with a complete, modular security
assessment framework. The new engine discovers Bluetooth targets through BlueZ, correlates
evidence, matches versioned target profiles, selects compatible modules, and produces
reproducible exploitability reports.

Important

PerfektBlue does not guess payload offsets, architectures, ports, or injection paths.
Active verification requires a tested target profile and compatible module. Unknown targets
are mapped and reported without fabricated exploit claims.

✨ Highlights

Area PerfektBlue 2.0
Bluetooth BlueZ D-Bus discovery for Classic Bluetooth and BLE
Fingerprinting Explainable weighted profile matching with contradictions
Intelligence Adaptive module planning based on live evidence and confidence
Assessment Precise injectable, blocked, inconclusive, and unsupported verdicts
Reporting Persistent sessions plus terminal, JSON, and standalone HTML reports
Safety Passive-by-default policy with explicit lab-active authorization
CAN Isolated as the optional perfektblue-can plugin
Distribution Python wheel, source archive, reproducible .deb, checksums, and SBOM

🔍 Bluetooth intelligence

  • Enumerates adapters, devices, service UUIDs, GATT services, characteristics, flags, handles,
    manufacturer data, RSSI, device class, and link-security state.
  • Builds a capability-aware assessment plan and explains why every module runs or is skipped.
  • Uses target-profile confidence thresholds before considering active verification.
  • Stores evidence and findings in an indexed session database for later reporting.

🧪 Verified assessment model

The framework now uses explicit verdict semantics:

  • confirmed-injectable — a compatible module produced its required proof.
  • blocked — a known path was tested and rejected by a mitigation.
  • no-known-path — the matched profile has no installed verified path.
  • inconclusive — evidence or authorization was insufficient.
  • unsupported — no target profile matched.

The release includes vulnerable, patched, and unknown simulators so profile authors can test all
verdict paths without real hardware.

📦 Installation

Download perfektblue_2.0.0-1_all.deb from the assets below:

sudo apt install ./perfektblue_2.0.0-1_all.deb
perfektblue doctor
perfektblue wizard

Python installation:

python3 -m venv .venv
. .venv/bin/activate
python -m pip install perfektblue-2.0.0-py3-none-any.whl
perfektblue doctor

✅ Release assurance

  • 26 automated tests
  • 81.4% branch-aware coverage
  • Ruff formatting and linting
  • Strict mypy validation
  • Dependency vulnerability audit
  • Reproducible Debian build verification
  • Clean wheel and Debian installation smoke tests
  • SHA-256 checksums
  • SPDX SBOM
  • GitHub build-provenance attestation

⚠️ Authorized use

Use PerfektBlue only on systems you own or have explicit written permission to assess. Automotive
and Bluetooth systems may be safety-critical. Use isolated labs, simulators, and controlled test
benches.

Full usage and architecture documentation is available in the repository
README.

Assets 9
Loading