norrisaftcc / norrisaftcc/algorithm-liza
Write compact variants of Kevin and the test engineer
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 0
- Forks
- 0
- Avg merge
- 1m
- Merged PRs (30d)
- 1
Description
The two personas that plausibly describe behaviour a small local model could perform on this project's task class: Kevin (GitHub process, issues, PRs) and the test engineer.
The compact variant is a rewrite, not a truncation. The imported prompts run 1,500-3,000 tokens; ADR-0002's budget is roughly a thousand for the entire system prompt. Most of what makes these prompts good for Claude is precisely the nuance a 4B model drops on the floor, and for a small model four numbered "how you work" steps outperform four paragraphs of character description.
Whether the compact variants are good enough is an eval question, not an opinion — so this is blocked on M3.
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 ADR-0002 and the imported Kevin and test engineer prompts. Check the M3 evaluation before rewriting them into compact variants within the roughly 1,000-token system-prompt budget. Done means both variants are available and the M3 evaluation shows they are good enough for the task class.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100