PRISM-GEN-DEMO

- No file changes detected in this release. - Documentation and user guide remain unchanged. - Functionality and interface are stable with no modifications.

Version 2.0.1 Major refactoring: Removed all Bash scripts, shell dependencies, and legacy pipeline code. Reimplemented the core as a fully self-contained, secure Python analysis suite. - All analysis functions are now Python scripts (retrieve, filter, sort, merge, score, plot, summary) under scripts/ - Data analysis is entirely offline and read-only; all required CSV data is bundled in data/ - No shell/bash scripting, subprocess calls, or hardcoded paths—pure Python throughout - No network activity, no API keys, no credential handling; outputs are written only to specified paths - Visualization (plotting) is now Python/matplotlib-based (optional) - Security and portability improved: greatly reduced codebase (51 files removed, 10 added), modernized architecture

**Version 1.0.8 Changelog** - Major structural expansion: 56 new files added, including all core analysis scripts, precomputed data, assets, documentation, and configuration. - Reorganized codebase with modules for each PRISM-Gen workflow step (VAE, surrogate, DFT, ADMET, docking, etc). - Added multiple pre-calculated CSV result files to the data directory for demonstration and analysis. - Extensive documentation added: new tutorial, detailed README, and usage guides for easier onboarding. - Removed old plotting scripts and version info to reduce redundancy and clarify workflow structure.

**Major codebase simplification and security update.** - Removed 50 maintenance, auxiliary, and legacy files, leaving only core analysis and plotting scripts. - Added three essential scripts for data preview and visualization: `demo_plot_distribution.py`, `demo_plot_scatter.py`, and `VERSION.md`. - All self-modifying, fix, packaging, and legacy scripts scrapped for maximum safety and transparency. - Skill now consists of Bash-only core analysis, with Python required only for advanced visualization. - All user input validated; no unsafe code patterns remain. - Fully aligned documentation and clear usage instructions; easy to audit and maintain.

Version 1.0.6 - Added multiple new scripts for improved security and execution: - Security-focused Python and shell scripts (e.g., complete_security_fix_v1.0.6.py, fix_all_security_v1.0.6.py, fix_v1.0.6_simple.sh, quick_fix_v1.0.6.sh). - New scripts under safe_scripts_v2/ for safe plotting and Python execution (safe_demo_plot_distribution.py, safe_demo_plot_scatter.py, safe_exec.py, safe_python_executor.py). - Enhanced overall security and execution safety for advanced functions and data analysis.

**v1.0.5 – Security-focused update with major codebase cleanup and safety enhancements** - Fixed critical RCE vulnerabilities by removing unsafe shell variable interpolation and implementing secure argument passing. - Deleted all self-modifying and path-dependent helper scripts (e.g., `_update_python_calls.sh`, `_fix_python_bool.sh`). - Added comprehensive input validation for all user and file inputs. - Replaced legacy shell scripts with secure Python wrappers in `safe_scripts/`; all command-line interfaces are now RCE-safe. - Updated documentation to reflect new security model and removed outdated/unsecured usage examples. - Project structure and metadata files streamlined for safety and future ClawHub compatibility.

- Major change: Core Python implementation removed; outputs now provided as pre-generated result files only. - Removed 12 core Python pipeline scripts from the project. - Added pre-filtered and top10 result CSVs under the output/ directory for direct data access. - Skill now functions entirely on pre-calculated CSV outputs for demonstration and querying. - No change to user-facing documentation or technical/environment requirements.

- Expanded core functionality with new pipeline and processing scripts under core/ - Added multiple CSV data files under data/ and demo/ for pre-computed molecule screening and analysis - Introduced offline-only mode for all basic functions, with Python dependencies now clearly optional for advanced visualization and analytics - Updated documentation to emphasize offline usage, distinguishing between basic (no install required) and advanced (Python required) features - Removed legacy documentation files (CONTRIBUTING.md, INSTALL.md, RELEASE_CHECKLIST.md, RELEASE_NOTES.md) - Refined technical requirements and usage guidance for clearer compatibility and setup steps

- Added full English descriptions and bilingual (English/Chinese) documentation throughout the SKILL.md, increasing international accessibility. - Updated title, description, tags, and categories to include both languages. - Reorganized feature sections in SKILL.md with clearer bilingual headings and concise bullet points. - Clarified supported CSV file names and mapped their descriptions for both languages. - Provided updated technical requirements and compatibility notes in both English and Chinese. - No functional or code changes; documentation updates only.

PRISM-Gen Demo 1.0.0 — Initial Release - Provides portable analysis of precomputed PRISM-Gen molecular CSV results - Supports flexible data querying: retrieval, filtering (complex/multi-condition), sorting, merging, and visualization - Includes rich command-line scripts for source management, search, ranking, cross-stage comparison, and plotting - Outputs results in a structured, user-friendly format suitable for further processing or export - Strong focus on data authenticity: only actual precomputed results are handled, never simulated or fake data - Extensible framework: custom data sources, rules, and plugins supported for advanced workflows