oscura 0.8.0__py3-none-any.whl → 0.10.0__py3-none-any.whl

{oscura-0.8.0.dist-info → oscura-0.10.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: oscura
-Version: 0.8.0
+Version: 0.10.0
 Summary: Unified hardware reverse engineering framework. Extract all information from any system through signals and data. Unknown protocol discovery, state machine extraction, CRC recovery, security analysis. 16+ protocols, IEEE-compliant measurements.
 Project-URL: Homepage, https://github.com/oscura-re/oscura
 Project-URL: Documentation, https://github.com/oscura-re/oscura/tree/main/docs
@@ -28,16 +28,26 @@ Classifier: Topic :: System :: Hardware
 Classifier: Topic :: System :: Hardware :: Hardware Drivers
 Classifier: Typing :: Typed
 Requires-Python: >=3.12
+Requires-Dist: cantools<40.0.0,>=39.4.0
 Requires-Dist: click<9.0.0,>=8.1.0
 Requires-Dist: dpkt<2.0.0,>=1.9.0
+Requires-Dist: jinja2<4.0.0,>=3.1
+Requires-Dist: matplotlib<4.0.0,>=3.7.0
 Requires-Dist: numpy<3.0.0,>=1.24.0
+Requires-Dist: openpyxl<4.0.0,>=3.0.0
+Requires-Dist: pandas<3.0.0,>=2.0.0
+Requires-Dist: psutil<7.0.0,>=5.9.0
+Requires-Dist: python-can<5.0.0,>=4.4.0
+Requires-Dist: python-pptx<1.0.0,>=0.6.21
 Requires-Dist: pyyaml<7.0.0,>=6.0
+Requires-Dist: reportlab<6.0.0,>=4.4.7
+Requires-Dist: scapy<3.0.0,>=2.5.0
 Requires-Dist: scipy<2.0.0,>=1.10.0
 Requires-Dist: tm-data-types<1.0.0,>=0.3.0
 Requires-Dist: tqdm<5.0.0,>=4.65.0
+Requires-Dist: weasyprint<64.0.0,>=63.0
 Provides-Extra: all
 Requires-Dist: asammdf<9.0.0,>=8.0.0; extra == 'all'
-Requires-Dist: cantools<40.0.0,>=39.4.0; extra == 'all'
 Requires-Dist: check-jsonschema<1.0.0,>=0.29.0; extra == 'all'
 Requires-Dist: coverage[toml]<8.0.0,>=7.0; extra == 'all'
 Requires-Dist: cryptography<47.0.0,>=44.0.1; extra == 'all'
@@ -45,17 +55,12 @@ Requires-Dist: h5py<4.0.0,>=3.0.0; extra == 'all'
 Requires-Dist: hypothesis<7.0.0,>=6.0.0; extra == 'all'
 Requires-Dist: interrogate<2.0.0,>=1.7.0; extra == 'all'
 Requires-Dist: ipython<9.0.0,>=8.0.0; extra == 'all'
-Requires-Dist: jinja2<4.0.0,>=3.1; extra == 'all'
 Requires-Dist: jupyter<2.0.0,>=1.0.0; extra == 'all'
-Requires-Dist: matplotlib<4.0.0,>=3.7.0; extra == 'all'
 Requires-Dist: mypy<2.0.0,>=1.0; extra == 'all'
 Requires-Dist: nbconvert!=7.16.6,<8.0.0,>=7.0.0; extra == 'all'
 Requires-Dist: networkx<4.0.0,>=3.0; extra == 'all'
 Requires-Dist: nptdms<2.0.0,>=1.7.0; extra == 'all'
-Requires-Dist: openpyxl<4.0.0,>=3.0.0; extra == 'all'
-Requires-Dist: pandas<3.0.0,>=2.0.0; extra == 'all'
 Requires-Dist: pre-commit<5.0.0,>=4.0.0; extra == 'all'
-Requires-Dist: psutil<7.0.0,>=5.9.0; extra == 'all'
 Requires-Dist: pyserial<4.0.0,>=3.5; extra == 'all'
 Requires-Dist: pytest-asyncio<1.0.0,>=0.23.0; extra == 'all'
 Requires-Dist: pytest-benchmark<6.0.0,>=4.0.0; extra == 'all'
@@ -66,16 +71,12 @@ Requires-Dist: pytest-split<1.0.0,>=0.10.0; extra == 'all'
 Requires-Dist: pytest-timeout<3.0.0,>=2.3.0; extra == 'all'
 Requires-Dist: pytest-xdist<4.0.0,>=3.0; extra == 'all'
 Requires-Dist: pytest<10.0.0,>=8.0; extra == 'all'
-Requires-Dist: python-can<5.0.0,>=4.4.0; extra == 'all'
-Requires-Dist: python-pptx<1.0.0,>=0.6.21; extra == 'all'
 Requires-Dist: pyvisa-py<1.0.0,>=0.7.0; extra == 'all'
 Requires-Dist: pyvisa<2.0.0,>=1.13.0; extra == 'all'
 Requires-Dist: pywavelets<2.0.0,>=1.0.0; extra == 'all'
 Requires-Dist: rapidfuzz<4.0.0,>=3.0.0; extra == 'all'
-Requires-Dist: reportlab<6.0.0,>=4.4.7; extra == 'all'
 Requires-Dist: rigolwfm<2.0.0,>=1.0.0; extra == 'all'
 Requires-Dist: ruff<1.0.0,>=0.8.0; extra == 'all'
-Requires-Dist: scapy<3.0.0,>=2.5.0; extra == 'all'
 Requires-Dist: scikit-learn<2.0.0,>=1.3.0; extra == 'all'
 Requires-Dist: types-pyyaml<7.0.0,>=6.0; extra == 'all'
 Requires-Dist: yamllint<2.0.0,>=1.35; extra == 'all'
@@ -85,12 +86,6 @@ Requires-Dist: pywavelets<2.0.0,>=1.0.0; extra == 'analysis'
 Requires-Dist: scikit-learn<2.0.0,>=1.3.0; extra == 'analysis'
 Provides-Extra: automotive
 Requires-Dist: asammdf<9.0.0,>=8.0.0; extra == 'automotive'
-Requires-Dist: cantools<40.0.0,>=39.4.0; extra == 'automotive'
-Requires-Dist: python-can<5.0.0,>=4.4.0; extra == 'automotive'
-Requires-Dist: scapy<3.0.0,>=2.5.0; extra == 'automotive'
-Provides-Extra: dataframes
-Requires-Dist: openpyxl<4.0.0,>=3.0.0; extra == 'dataframes'
-Requires-Dist: pandas<3.0.0,>=2.0.0; extra == 'dataframes'
 Provides-Extra: dev
 Requires-Dist: check-jsonschema<1.0.0,>=0.29.0; extra == 'dev'
 Requires-Dist: coverage[toml]<8.0.0,>=7.0; extra == 'dev'
@@ -127,28 +122,12 @@ Requires-Dist: nbconvert!=7.16.6,<8.0.0,>=7.0.0; extra == 'jupyter'
 Provides-Extra: oscilloscopes
 Requires-Dist: nptdms<2.0.0,>=1.7.0; extra == 'oscilloscopes'
 Requires-Dist: rigolwfm<2.0.0,>=1.0.0; extra == 'oscilloscopes'
-Provides-Extra: reporting
-Requires-Dist: jinja2<4.0.0,>=3.1; extra == 'reporting'
-Requires-Dist: matplotlib<4.0.0,>=3.7.0; extra == 'reporting'
-Requires-Dist: python-pptx<1.0.0,>=0.6.21; extra == 'reporting'
-Requires-Dist: reportlab<6.0.0,>=4.4.7; extra == 'reporting'
 Provides-Extra: standard
-Requires-Dist: jinja2<4.0.0,>=3.1; extra == 'standard'
-Requires-Dist: matplotlib<4.0.0,>=3.7.0; extra == 'standard'
-Requires-Dist: openpyxl<4.0.0,>=3.0.0; extra == 'standard'
-Requires-Dist: pandas<3.0.0,>=2.0.0; extra == 'standard'
-Requires-Dist: psutil<7.0.0,>=5.9.0; extra == 'standard'
-Requires-Dist: python-pptx<1.0.0,>=0.6.21; extra == 'standard'
-Requires-Dist: reportlab<6.0.0,>=4.4.7; extra == 'standard'
-Provides-Extra: system
-Requires-Dist: psutil<7.0.0,>=5.9.0; extra == 'system'
-Provides-Extra: visualization
-Requires-Dist: matplotlib<4.0.0,>=3.7.0; extra == 'visualization'
 Description-Content-Type: text/markdown
 
 # Oscura
 
-**Workflow automation for hardware reverse engineering.** Stop juggling seven different tools to analyze one capture. Oscura chains specialized tools (sigrok, ChipWhisperer, scipy) into unified Python workflows—from oscilloscope files to Wireshark dissectors without manual conversions or context switching.
+**Workflow automation for hardware reverse engineering.** Stop juggling seven different tools to analyze one capture. Oscura provides unified Python workflows—from oscilloscope files to Wireshark dissectors without manual conversions or context switching.
 
 [![CI](https://github.com/oscura-re/oscura/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/oscura-re/oscura/actions/workflows/ci.yml)
 [![Code Quality](https://github.com/oscura-re/oscura/actions/workflows/code-quality.yml/badge.svg?branch=main)](https://github.com/oscura-re/oscura/actions/workflows/code-quality.yml)
@@ -164,8 +143,8 @@ Description-Content-Type: text/markdown
 Hardware reverse engineering means juggling specialized tools:
 
 1. Export oscilloscope waveforms (vendor-specific formats)
-2. Convert formats for analysis (sigrok, custom scripts)
-3. Decode protocols (PulseView, separate decoders)
+2. Convert formats for analysis (custom scripts, manual conversion)
+3. Decode protocols (separate decoder tools)
 4. Infer unknown protocols (Netzob, manual analysis)
 5. Reverse checksums (CRC RevEng, separate tool)
 6. Generate documentation (manual Wireshark dissectors, DBC files)
@@ -177,27 +156,23 @@ Hardware reverse engineering means juggling specialized tools:
 
 Oscura automates complete workflows in Python:
 
-**What We Integrate:**
-
-- Protocol decoding via [sigrok](https://sigrok.org/) (UART, SPI, I2C, CAN, etc.)
-- Signal processing with scipy/numpy
-- Side-channel trace formats (ChipWhisperer)
-- Automotive protocols (cantools integration)
-
-**What We Add:**
+**What We Provide:**
 
+- **Native protocol decoders for 16 formats**: UART, SPI, I2C, CAN, CAN-FD, LIN, FlexRay, JTAG, SWD, USB, I2S, 1-Wire, HDLC, Manchester
 - **Hypothesis-driven RE workflows** with differential analysis and confidence scoring
 - **Automatic Wireshark dissector generation** from inferred protocols
 - **DBC file generation** from raw CAN captures (no manual signal definition)
-- **Multi-format file loading** (Tektronix, Rigol, Sigrok, BLF, PCAP, ChipWhisperer)
+- **Multi-format file loading** (Tektronix, Rigol, Sigrok .sr, BLF, PCAP, ChipWhisperer, VCD, WAV, CSV, HDF5)
 - **CRC/checksum recovery** from message-checksum pairs
 - **Unified Python API** eliminating tool-hopping and format conversions
+- Signal processing with scipy/numpy
+- Automotive protocol support (cantools integration)
 
 **Value proposition:** Write one Python script instead of:
 
 1. Exporting from oscilloscope software (vendor GUI)
-2. Converting formats (sigrok-cli, custom scripts)
-3. Decoding protocols (PulseView manual selection)
+2. Converting formats (custom scripts, manual conversion)
+3. Decoding protocols (separate decoder tool manual configuration)
 4. Inferring message formats (Netzob or manual)
 5. Recovering checksums (CRC RevEng separate invocation)
 6. Writing dissectors (manual Lua coding)
@@ -292,38 +267,41 @@ result = osc.auto_decode(trace)
 print(f"Detected {result.protocol}: {len(result.frames)} frames decoded")
 ```
 
-**[See demos/README.md](demos/README.md)** for 105+ comprehensive demonstrations organized by skill level.
+**[See demos/README.md](demos/README.md)** for 100+ working demonstrations organized by skill level (36 comprehensive workflows across 19 categories).
 
 ---
 
 ## Core Capabilities
 
-### Unique Contributions
+### What Oscura Does Uniquely
 
 | Capability                            | What We Provide                                                                                   | Why It Matters                                                                           |
 | ------------------------------------- | ------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------- |
 | **Hypothesis-Driven RE**              | BlackBoxSession with differential analysis, field detection, confidence scoring, audit trails     | Systematic unknown protocol analysis vs manual guesswork                                 |
 | **DBC Auto-Generation**               | Statistical CAN signal inference from captures → DBC export                                       | Open-source alternative to Vector CANalyzer ($$$)                                        |
 | **Wireshark Dissector Generation**    | Infer protocol → generate validated Lua dissector                                                 | End-to-end automation (others require manual YAML specs)                                 |
-| **Multi-Format File Loading**         | Oscilloscopes (Tektronix WFM, Rigol), logic analyzers (Sigrok, VCD), automotive BLF  | Eliminate format conversion steps                                                        |
-| **Statistical Protocol Auto-Detect**  | Waveform analysis (timing, voltage, patterns) → protocol identification                           | Goes beyond sigrok's signal name matching                                                |
+| **Multi-Format File Loading**         | 22 format loaders: oscilloscopes (Tektronix WFM, Rigol), logic analyzers (Sigrok .sr, VCD), automotive (BLF, MF4), ChipWhisperer, PCAP, WAV, CSV, HDF5  | Eliminate format conversion steps                                                        |
+| **Statistical Protocol Auto-Detect**  | Waveform analysis (timing, voltage, patterns) → protocol identification                           | Automatic protocol detection without manual configuration                                |
 | **Unified Workflow API**              | Single Python script: oscilloscope file → decode → infer → export dissector                       | Replace 7-tool chains with one script                                                    |
-| **CRC Recovery**                      | Message-checksum pairs → polynomial, init, XOR out, reflection                                    | Practical automation (CRC RevEng is more robust for edge cases)                          |
-| **Automotive Security Analysis**      | Stimulus-response correlation, hypothesis testing, UDS/OBD-II decoding                            | Research-focused (CANToolz covers security, python-can covers low-level)                 |
-| **State Machine Extraction (Passive)** | RPNI algorithm for passive observation (vs Netzob's active L\* requiring oracle)                  | Different use case from existing tools                                                   |
+| **CRC Recovery**                      | Message-checksum pairs → polynomial, init, XOR out, reflection                                    | Practical automation for common cases                                                    |
+| **Automotive Security Analysis**      | Stimulus-response correlation, hypothesis testing, UDS/OBD-II decoding                            | Research-focused workflow automation                                                     |
+| **State Machine Extraction (Passive)** | RPNI algorithm for passive observation (vs active learning requiring oracle)                      | Different use case from Netzob's active L* algorithm                                     |
 | **Evidence-Based Discovery**          | Confidence scoring, hypothesis tracking, statistical validation, reproducible audit trails        | Scientific rigor for research publication                                                |
 
-### Integration Capabilities
+### When to Use Alternatives
+
+Oscura automates workflows but isn't always the best tool for every job. Use specialized tools when appropriate:
 
-| Category                   | Implementation                                                                          | Best Alternative                                                    |
-| -------------------------- | --------------------------------------------------------------------------------------- | ------------------------------------------------------------------- |
-| **Protocol Decoding**      | Integrated sigrok decoders (UART, SPI, I2C, CAN, LIN, JTAG, etc.) via Python API       | [sigrok](https://sigrok.org/) directly (100+ decoders)              |
-| **Side-Channel Analysis**  | Load ChipWhisperer traces, basic DPA/CPA implementations                               | [ChipWhisperer](https://chipwhisperer.com/) (superior capabilities) |
-| **Signal Processing**      | IEEE-based measurements using scipy/numpy                                               | [scipy.signal](https://docs.scipy.org/doc/scipy/reference/signal.html) directly or MATLAB                        |
-| **CAN Parsing**            | cantools integration for DBC parsing and message encoding                               | [cantools](https://github.com/cantools/cantools) + [python-can](https://python-can.readthedocs.io/)     |
-| **File Format Conversion** | Loaders for 13+ formats with unified API                                                | Vendor software + manual export                                     |
+| Need                       | Oscura Capability                                      | When to Use Alternative Instead                                     |
+| -------------------------- | ------------------------------------------------------ | ------------------------------------------------------------------- |
+| **Protocol Decoding**      | Native decoders for 16 protocols (UART, SPI, I2C, CAN, LIN, FlexRay, JTAG, SWD, USB, I2S, 1-Wire, HDLC, Manchester, CAN-FD, OneWire) | **Use [sigrok](https://sigrok.org/)** if you need 100+ protocol decoders, real-time hardware acquisition, or mature PulseView GUI              |
+| **Side-Channel Analysis**  | Demo-level DPA/CPA, ChipWhisperer trace loading, basic power analysis                               | **Use [ChipWhisperer](https://chipwhisperer.com/)** for production side-channel attacks—superior capabilities, hardware integration, mature attack implementations |
+| **Signal Processing**      | IEEE-based measurements using scipy/numpy                                               | **Use [scipy.signal](https://docs.scipy.org/doc/scipy/reference/signal.html) or MATLAB** for advanced DSP, optimized performance, comprehensive filter design                        |
+| **CRC Recovery**           | Message-checksum pairs → CRC spec (common cases)       | **Use [CRC RevEng](https://reveng.sourceforge.io/)** for edge cases, exhaustive search, or when our heuristics fail |
+| **CAN Bus Analysis**            | DBC generation, cantools integration, automotive protocol support                               | **Use [python-can](https://python-can.readthedocs.io/)** for low-level CAN interface control, or **[CANalyzer](https://www.vector.com/int/en/products/products-a-z/software/canalyzer/)** for professional automotive work     |
+| **File Format Conversion** | Loaders for 22 formats with unified API                                                | **Use vendor software** for proprietary formats we don't support, or when you need 100% format fidelity                                     |
 
-**Our philosophy:** Integrate best-in-class tools rather than reimplementing them. Add value through workflow automation and novel analysis methods.
+**Oscura's sweet spot:** Chaining multiple RE steps in scripted workflows with hypothesis-driven analysis. If you only need one specialized capability, use the dedicated tool.
 
 ---
 
@@ -337,16 +315,18 @@ print(f"Detected {result.protocol}: {len(result.frames)} frames decoded")
 - You need Wireshark dissectors generated automatically from inferred protocols
 - You're working with multiple oscilloscope/LA formats and want unified API
 - You value reproducible research with hypothesis tracking and confidence scoring
+- You want to chain multiple RE tools together in one script
 
 **Use specialized tools directly when:**
 
-- You only need protocol decoding → [sigrok](https://sigrok.org/) has 100+ decoders
-- You're doing side-channel attacks → [ChipWhisperer](https://chipwhisperer.com/) is superior
-- You only need signal processing → [scipy](https://scipy.org/)/MATLAB are more optimized
+- You only need protocol decoding → [sigrok](https://sigrok.org/) has 100+ decoders and hardware support
+- You're doing production side-channel attacks → [ChipWhisperer](https://chipwhisperer.com/) has superior capabilities
+- You only need signal processing → [scipy](https://scipy.org/)/MATLAB are more comprehensive and optimized
 - You need the most robust CRC recovery → [CRC RevEng](https://reveng.sourceforge.io/) handles edge cases better
 - You have vendor-specific needs → vendor tools have more format support
+- You need real-time hardware acquisition → sigrok, vendor tools, or python-can have better hardware integration
 
-**Oscura's sweet spot:** Chaining multiple RE steps in scripted workflows with novel hypothesis-driven analysis.
+**Oscura's sweet spot:** Workflow automation that chains multiple RE steps with hypothesis-driven analysis.
 
 ---
 
@@ -357,7 +337,7 @@ print(f"Detected {result.protocol}: {len(result.frames)} frames decoded")
 - **Protocol reverse engineering** with hypothesis tracking and validation
 - **Automotive ECU security** via CAN stimulus-response analysis
 - **Attack surface mapping** through state machine extraction
-- **Cryptographic implementation validation** (use ChipWhisperer for attacks, Oscura for trace analysis workflows)
+- **Side-channel trace analysis workflows** (demo-level DPA/CPA—use ChipWhisperer for production attacks)
 
 ### Right-to-Repair & Modernization
 
@@ -371,7 +351,7 @@ print(f"Detected {result.protocol}: {len(result.frames)} frames decoded")
 - **Reproducible workflows** with evidence tracking and audit trails
 - **Statistical validation** with confidence scoring
 - **IEEE-based measurements** for publishable results (181/1241/1459/2414)
-- **22,000+ comprehensive tests, 80%+ coverage** ensure reliability
+- **475 test files with 80%+ coverage** ensure reliability
 
 ### Industrial & Automotive
 
@@ -384,18 +364,18 @@ print(f"Detected {result.protocol}: {len(result.frames)} frames decoded")
 
 ## Built On
 
-Oscura integrates proven open-source tools:
+Oscura builds on proven open-source tools:
 
 | Component           | What We Use                                                             | Why                                                     |
 | ------------------- | ----------------------------------------------------------------------- | ------------------------------------------------------- |
-| **Protocol Engine** | [sigrok](https://sigrok.org/) libsigrokdecode                           | 100+ mature, community-supported protocol decoders      |
 | **Signal Processing** | [scipy](https://scipy.org/)/[numpy](https://numpy.org/)                 | Industry-standard numerical computing                   |
-| **Side-Channel Traces** | [ChipWhisperer](https://chipwhisperer.com/) formats                     | De facto standard for side-channel research             |
+| **Protocol Decoders** | Native Python implementations (16 protocols)                           | Full control, Python integration, no external dependencies |
+| **File Format Loaders** | Custom parsers for 22 formats (WFM, Rigol, .sr, BLF, PCAP, etc.)      | Support oscilloscopes, logic analyzers, automotive, ChipWhisperer |
 | **CAN Protocols**   | [cantools](https://github.com/cantools/cantools), [python-can](https://python-can.readthedocs.io/) | Robust CAN message parsing and encoding                |
 | **Testing**         | [pytest](https://pytest.org/), [Hypothesis](https://hypothesis.readthedocs.io/)         | Property-based testing for algorithm validation         |
 | **Type Safety**     | [mypy](https://mypy-lang.org/)                                          | Static type checking (strict mode)                      |
 
-**Our contribution:** Unified API + novel hypothesis-driven RE workflows + format handling + export automation.
+**Our contribution:** Unified API + hypothesis-driven RE workflows + 22 format loaders + 16 native protocol decoders + export automation (Wireshark, DBC, Scapy).
 
 ---
 
@@ -405,8 +385,8 @@ Oscura integrates proven open-source tools:
 
 Production-ready validation:
 
-- **22,000+ comprehensive tests** with property-based validation (Hypothesis)
-- **80%+ code coverage** with branch coverage enabled
+- **475 test files** with comprehensive unit, integration, and property-based tests (Hypothesis)
+- **80%+ code coverage** enforced with branch coverage enabled
 - **Pre-commit hooks** (format, lint, type check) enforce consistency
 - **Merge queue CI** prevents untested code from landing
 - **Nightly stress tests** validate edge cases and memory usage
@@ -441,7 +421,7 @@ Built for extensibility:
 
 ### Working Demonstrations
 
-**105+ comprehensive demos** organized into 12 categories covering:
+**100+ working demos** organized into 19 categories covering:
 
 - **Data Loading** - All file format loaders (oscilloscopes, logic analyzers, automotive, scientific)
 - **Basic Analysis** - Waveform measurements, digital analysis, spectral analysis, filtering
@@ -458,7 +438,7 @@ Built for extensibility:
 
 ### Comprehensive Demonstrations
 
-**33+ in-depth demos** organized by skill level and domain:
+**36 comprehensive workflows** organized by skill level and domain:
 
 - **[Getting Started](demos/README.md#beginner-path-2-4-hours)** - File loading, basic measurements, format export (Beginner, 2-4 hours)
 - **[Protocol Decoding](demos/README.md#intermediate-path-6-10-hours)** - UART, SPI, I2C, Manchester, JTAG, USB, PCAP (Intermediate, 6-10 hours)
@@ -610,7 +590,7 @@ python3 .claude/hooks/validate_all.py # Must show 5/5 passing
 
 ## Project Status
 
-**Current Version:** [0.6.0](https://github.com/oscura-re/oscura/releases/latest) (2026-01-25)
+**Current Version:** [0.9.0](https://github.com/oscura-re/oscura/releases/latest) (2026-01-31)
 
 **Active Development Areas:**
 
@@ -636,7 +616,7 @@ If Oscura contributes to your research, please cite:
   author = {Oscura Contributors},
   year = {2026},
   url = {https://github.com/oscura-re/oscura},
-  version = {0.6.0}
+  version = {0.9.0}
 }
 ```