prompt-database 0.1.0__tar.gz

prompt_database-0.1.0/.editorconfig

@@ -0,0 +1,25 @@
+# EditorConfig helps maintain consistent coding styles
+# https://editorconfig.org
+
+root = true
+
+[*]
+charset = utf-8
+end_of_line = lf
+insert_final_newline = true
+trim_trailing_whitespace = true
+
+[*.{py,sh}]
+indent_style = space
+indent_size = 4
+
+[*.{json,yml,yaml}]
+indent_style = space
+indent_size = 2
+
+[*.md]
+trim_trailing_whitespace = false
+max_line_length = off
+
+[Makefile]
+indent_style = tab

prompt_database-0.1.0/.gitattributes

@@ -0,0 +1,29 @@
+# Auto detect text files and perform LF normalization
+* text=auto
+
+# Source code
+*.py text eol=lf
+*.sh text eol=lf
+*.md text eol=lf
+*.txt text eol=lf
+*.json text eol=lf
+*.yml text eol=lf
+*.yaml text eol=lf
+
+# Git LFS - Track large files
+*.db filter=lfs diff=lfs merge=lfs -text
+*.sqlite filter=lfs diff=lfs merge=lfs -text
+*.sqlite3 filter=lfs diff=lfs merge=lfs -text
+
+# Large JSON files (if you decide to commit them)
+curated_advanced_prompts*.json filter=lfs diff=lfs merge=lfs -text
+
+# Binary files
+*.pkl binary
+*.pickle binary
+*.npy binary
+*.npz binary
+*.pt binary
+*.pth binary
+*.h5 binary
+*.hdf5 binary

prompt_database-0.1.0/.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,29 @@
+---
+name: Bug Report
+about: Report a bug or issue
+title: '[BUG] '
+labels: bug
+assignees: ''
+---
+
+## Bug Description
+<!-- A clear description of the bug -->
+
+## Steps to Reproduce
+1. 
+2. 
+3. 
+
+## Expected Behavior
+<!-- What you expected to happen -->
+
+## Actual Behavior
+<!-- What actually happened -->
+
+## Environment
+- OS: 
+- Python version: 
+- Database version: 
+
+## Additional Context
+<!-- Screenshots, logs, or other relevant information -->

prompt_database-0.1.0/.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,22 @@
+---
+name: Feature Request
+about: Suggest a new feature or improvement
+title: '[FEATURE] '
+labels: enhancement
+assignees: ''
+---
+
+## Feature Description
+<!-- Clear description of the proposed feature -->
+
+## Use Case
+<!-- Why is this feature needed? What problem does it solve? -->
+
+## Proposed Solution
+<!-- How should this feature work? -->
+
+## Alternatives Considered
+<!-- Other approaches you've thought about -->
+
+## Additional Context
+<!-- Any other relevant information -->

prompt_database-0.1.0/.github/ISSUE_TEMPLATE/prompt_submission.md

@@ -0,0 +1,49 @@
+---
+name: Prompt Submission
+about: Submit new attack prompts for the database
+title: '[PROMPT] '
+labels: prompt-addition
+assignees: ''
+---
+
+## Prompt Information
+
+**Attack Technique:**
+<!-- e.g., Prompt Extraction, Jailbreak, etc. -->
+
+**Complexity Level:**
+<!-- basic | intermediate | advanced -->
+
+**Attack Vector:**
+<!-- Direct | Indirect | Hybrid -->
+
+## Prompt Content
+```
+[Paste the attack prompt here]
+```
+
+## Description
+<!-- How this attack works, what it attempts to do -->
+
+## Test Results
+<!-- Which models did you test this against? What were the results? -->
+
+**Tested Models:**
+- [ ] Claude Sonnet 4.5
+- [ ] GPT-4
+- [ ] Other (specify):
+
+**Success Rate:** 
+
+**Sample Response:**
+```
+[If applicable, paste a redacted example of the model's response]
+```
+
+## Attribution
+<!-- If this is based on published research, provide citation/link -->
+
+## Responsible Use Declaration
+- [ ] I confirm this submission is for defensive security research only
+- [ ] I have not tested this against production systems without authorization
+- [ ] This prompt does not contain sensitive/proprietary information

prompt_database-0.1.0/.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,28 @@
+## Description
+<!-- Briefly describe what this PR does -->
+
+## Type of Change
+- [ ] Bug fix
+- [ ] New prompt additions
+- [ ] Database improvements
+- [ ] Documentation update
+- [ ] Other (please describe):
+
+## Testing
+<!-- Describe how you tested your changes -->
+
+- [ ] Tested prompts against multiple models
+- [ ] Verified database integrity
+- [ ] Checked for duplicates
+- [ ] Validated categorization
+
+## Checklist
+- [ ] Code follows project style guidelines
+- [ ] Comments added for complex logic
+- [ ] Documentation updated (if needed)
+- [ ] No sensitive data included
+- [ ] Commits follow conventional format
+- [ ] Changes are for defensive research purposes only
+
+## Additional Context
+<!-- Any other relevant information -->

prompt_database-0.1.0/.github/workflows/ci.yml

@@ -0,0 +1,40 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.11", "3.12", "3.13"]
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          lfs: true
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install dependencies
+        run: pip install -e ".[dev]"
+
+      - name: Lint with ruff
+        run: |
+          ruff check src/ tests/
+          ruff format --check src/ tests/
+
+      - name: Run tests
+        run: pytest tests/ -v --tb=short
+
+      - name: Verify build command
+        run: |
+          prompt-db build --data-dir . --output /tmp/ci_test.db --force
+          prompt-db --db /tmp/ci_test.db stats

prompt_database-0.1.0/.github/workflows/publish.yml

@@ -0,0 +1,32 @@
+name: Publish to PyPI
+
+on:
+  push:
+    tags:
+      - "v*"
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      id-token: write  # trusted publishing
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          lfs: true
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install build tools
+        run: pip install hatchling build
+
+      - name: Build package
+        run: python -m build
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1

prompt_database-0.1.0/.github/workflows/validate-prompts.yml

@@ -0,0 +1,78 @@
+name: Validate Prompt Submissions
+
+on:
+  pull_request:
+    paths:
+      - "submissions/**"
+      - "*.jsonl"
+
+jobs:
+  validate:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: write
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          lfs: true
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install prompt-database
+        run: pip install -e .
+
+      - name: Build reference database
+        run: prompt-db build --data-dir . --output /tmp/reference.db --force
+
+      - name: Find submission files
+        id: find-files
+        run: |
+          # Find new/changed JSONL or text files in submissions/ or root
+          FILES=$(git diff --name-only --diff-filter=ACM origin/main... -- 'submissions/*.jsonl' 'submissions/*.txt' '*.jsonl' | head -20)
+          echo "files=$FILES" >> "$GITHUB_OUTPUT"
+          if [ -z "$FILES" ]; then
+            echo "No submission files found"
+            echo "found=false" >> "$GITHUB_OUTPUT"
+          else
+            echo "found=true" >> "$GITHUB_OUTPUT"
+            echo "Found files: $FILES"
+          fi
+
+      - name: Validate submissions
+        if: steps.find-files.outputs.found == 'true'
+        run: |
+          EXIT=0
+          for file in ${{ steps.find-files.outputs.files }}; do
+            echo "=== Validating: $file ==="
+            prompt-db --db /tmp/reference.db validate "$file" --check-dupes || EXIT=1
+          done
+          exit $EXIT
+
+      - name: Post validation summary
+        if: always() && steps.find-files.outputs.found == 'true'
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const body = `### Prompt Submission Validation
+
+            The submission validation workflow ran on this PR.
+            Check the [Actions log](${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}) for details.
+
+            **What's checked:**
+            - Minimum content length (10+ chars)
+            - Attack pattern detection (60+ regex patterns)
+            - Quality scoring (must score 15+/100)
+            - Duplicate detection against existing database
+            `;
+
+            github.rest.issues.createComment({
+              issue_number: context.issue.number,
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              body: body
+            });

prompt_database-0.1.0/.gitignore

@@ -0,0 +1,89 @@
+# Database files (built from JSON sources via `prompt-db build`)
+*.db
+*.db-journal
+*.db-wal
+*.db-shm
+*.sqlite
+*.sqlite3
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments
+venv/
+env/
+ENV/
+.venv
+
+# IDEs
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+.DS_Store
+
+# Logs
+*.log
+logs/
+*.log.*
+
+# Temporary files
+tmp/
+temp/
+*.tmp
+*.bak
+*.backup
+
+# Research data (comment out if you want to commit datasets)
+prompt_injection_datasets/
+extracted_prompts/
+raw_data/
+datasets/
+
+# API keys and secrets
+.env
+.env.local
+*.key
+*.pem
+credentials.json
+config.local.*
+
+# OS files
+Thumbs.db
+.DS_Store
+desktop.ini
+
+# Jupyter
+.ipynb_checkpoints/
+*.ipynb_checkpoints
+
+# Model files (often too large for git)
+*.pt
+*.pth
+*.ckpt
+*.safetensors
+*.bin
+models/
+
+# Large JSON files (optional - uncomment if too large)
+# extracted_prompts.json

prompt_database-0.1.0/CHANGELOG.md

@@ -0,0 +1,32 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.1.0] - 2026-03-30
+
+### Added
+- **Python package** (`prompt-database`) installable via `pip install -e .`
+- **`prompt-db` CLI** with commands: `build`, `stats`, `search`, `export`, `info`, `audit`, `curate`
+- **SQLite schema** with FTS5 full-text search, SHA256 content-hash deduplication, and schema versioning
+- **OWASP LLM Top 10 (2025)** category mapping with correct descriptions
+- **MITRE ATLAS** technique IDs on categories for threat model interoperability
+- **Quality scoring engine** with 60+ regex patterns for identifying real attacks vs. noise
+- **Data curation pipeline** — audit and remove non-attack content (removes ~67% noise)
+- **Ingestion pipeline** for `curated_advanced_prompts_v2.json` and `elite_custom_prompts.json`
+- **Test result tracking** with automatic success_rate aggregation
+- **19 passing tests** covering schema, CRUD, search, dedup, quality, and build
+- **Export** to JSON, JSONL, and CSV formats
+
+### Changed
+- Database is now built from JSON sources via `prompt-db build` (no longer committed as binary)
+- Deduplication reduced 8,568 records to 3,983 unique prompts
+- Quality curation further reduces to ~1,300 high-signal attack prompts
+
+### Removed
+- Binary `prompts.db` from git tracking (build it yourself from JSON sources)
+- Phantom file references in README (db_manager.py, schema.sql, etc. that never existed)
+
+[0.1.0]: https://github.com/scthornton/prompt-database/releases/tag/v0.1.0

prompt_database-0.1.0/CONTRIBUTING.md

@@ -0,0 +1,68 @@
+# Contributing to Prompt Database
+
+Thank you for your interest in contributing to this defensive security research project!
+
+## Code of Conduct
+
+This project is for **defensive security research only**. All contributions must:
+- Focus on improving security defenses
+- Not enable or encourage malicious use
+- Comply with responsible disclosure practices
+
+## How to Contribute
+
+### Reporting Issues
+- Use GitHub Issues for bug reports and feature requests
+- Provide clear reproduction steps
+- Include relevant context and examples
+
+### Contributing Prompts
+When adding new attack prompts to the database:
+
+1. **Quality over quantity** - Focus on sophisticated, novel attacks
+2. **Categorization** - Properly tag with attack technique and complexity
+3. **Documentation** - Explain the attack mechanism and expected behavior
+4. **Testing** - Verify the prompt works as described
+5. **Attribution** - Credit original sources when applicable
+
+### Contribution Process
+
+1. Fork the repository
+2. Create a feature branch (`git checkout -b feature/your-feature`)
+3. Make your changes
+4. Test thoroughly
+5. Commit with clear messages (`git commit -m 'Add: sophisticated context manipulation technique'`)
+6. Push to your fork (`git push origin feature/your-feature`)
+7. Open a Pull Request
+
+### Commit Message Format
+
+```
+Type: Brief description
+
+Longer explanation if needed.
+
+- Bullet points for details
+- Reference issues: #123
+```
+
+Types: `Add`, `Fix`, `Update`, `Refactor`, `Docs`, `Test`
+
+### Code Style
+- Follow PEP 8 for Python code
+- Use meaningful variable names
+- Add comments for complex logic
+- Keep functions focused and testable
+
+### Testing
+- Test prompts against multiple models when possible
+- Document success rates and model responses
+- Report findings responsibly
+
+## Questions?
+
+Open an issue or reach out to the maintainers.
+
+## License
+
+By contributing, you agree that your contributions will be licensed under the MIT License.

prompt_database-0.1.0/LICENSE

@@ -0,0 +1,38 @@
+MIT License
+
+Copyright (c) 2025 Scott Thornton
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+---
+
+RESPONSIBLE USE NOTICE
+
+This database contains prompt injection and adversarial attack examples for 
+DEFENSIVE SECURITY RESEARCH ONLY. By using this software, you agree to:
+
+1. Use these materials only for legitimate security research, testing, and 
+   defense development purposes
+2. Not use these materials to attack, compromise, or harm any systems without 
+   explicit authorization
+3. Comply with all applicable laws and regulations
+4. Not weaponize or distribute these materials for malicious purposes
+
+The authors and contributors are not responsible for any misuse of this 
+database or the materials contained within it.

prompt_database-0.1.0/Makefile

@@ -0,0 +1,36 @@
+.PHONY: install dev test lint format build curate stats clean help
+
+help: ## Show this help
+	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-15s\033[0m %s\n", $$1, $$2}'
+
+install: ## Install the package
+	pip install -e .
+
+dev: ## Install with dev dependencies
+	pip install -e ".[dev]"
+
+test: ## Run tests
+	pytest tests/ -v
+
+lint: ## Run linter
+	ruff check src/ tests/
+	ruff format --check src/ tests/
+
+format: ## Auto-format code
+	ruff format src/ tests/
+	ruff check --fix src/ tests/
+
+build: ## Build the database from JSON sources
+	prompt-db build --data-dir . --output prompts.db --force
+
+curate: build ## Build and curate (remove noise)
+	prompt-db --db prompts.db curate
+
+stats: ## Show database statistics (build first if needed)
+	@test -f prompts.db || $(MAKE) build
+	prompt-db --db prompts.db stats
+
+clean: ## Remove generated files
+	rm -f prompts.db prompts.db-wal prompts.db-shm
+	rm -rf __pycache__ .pytest_cache .ruff_cache
+	find . -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true