NVIDIA-NeMo / NVIDIA-NeMo/Anonymizer
feat: move prompts to versioned template files
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 122
- Forks
- 17
- Avg merge
- 4d 2h
- Merged PRs (30d)
- 11
Description
Priority Level
Medium
Task Summary
Detection and replacement prompts are currently inline Python strings in detection_workflow.py and llm_replace_workflow.py. Move them to versioned template files for easier review, diffing, and iteration without touching Python code.
Technical Details & Implementation Plan
Create src/anonymizer/engine/prompts/ directory with template files (e.g. validation.txt, augmentation.txt, latent_detection.txt, replacement.txt). Load at module level or lazily. Keep <<PLACEHOLDER>> tag convention for variable substitution. Consider whether Jinja blocks should stay inline or move to the template files.
Suggested by @andreatgretel in PR #3 review.
Dependencies
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the inline prompts in detection_workflow.py and llm_replace_workflow.py, then inspect the surrounding src/anonymizer/engine structure. Compare how each workflow performs placeholder substitution and handles Jinja blocks. Done means the prompts are in versioned files under src/anonymizer/engine/prompts/ and both workflows preserve their existing behavior without inline prompt strings.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- backend
- Issue type
- Refactor
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100