Back to .md Directory
proj2rubric
Scores a project against a rubric with 30 criteria, each linked to evidence in the original repository.
ai rag workflow
View sourceWhat this file does
Scores a project against a rubric with 30 criteria, each linked to evidence in the original repository.
When to use it
- Evaluating a team project for a course or competition
- Auditing your own project against a standard checklist
- Creating a self-assessment template for open-source contributions
Total Score:
| Notes | Score | Evidence |
|---|---|---|
| Video1 | 3 | V2 LiveDemo |
| Bonus: Xfold improvement | 3 | Scalability |
| Docs: what: point descriptions of each class/function (in isolation) | 3 | Pdoc Generated GitHub |
| Use of style checkers | 3 | Code Linting Integrated GitHub Action |
| Use of code formatters. | 3 | Code Linting Integrated GitHub Action |
| Use of syntax checkers. | 3 | Code Linting Integrated GitHub Action |
| Use of code coverage | 3 | Badge in README.md |
| Other automated analysis tools | 3 | Evidence |
| Workload is spread over the whole team (one team member is often Xtimes more productive than the others, but nevertheless, here is a track record that everyone is contributing a lot | 3 | From GH Insights, team members have equally participated in scaling the original project |
| Number of commits | 3 | From Github commits |
| Number of commits: by different people | 3 | From GH Insights |
| Issues reports: there are many | 3 | Overall 10+ issues listed |
| Issues are being closed | 3 | At the time of submission, all issues have sparked discussions with teammates and have been closed. |
| DOI badge: exists | 3 | From GH within README.md |
| Docs: doco generated, format not ugly | 3 | Evidence |
| Docs: how: for common use cases X,Y,Z mini-tutorials showing worked examples on how to do X,Y,Z | 3 | LiveDemo |
| Docs: why: docs tell a story, motivate the whole thing, deliver a punchline that makes you want to rush out and use the thing | 3 | "Generate documents with Python and get the job done faster than ever before! With Python document generator, you can create professional-looking documents in no time, so you can focus on the important things. Get ready to revolutionize your workflow - it's time to take your productivity to the next level!" |
| Docs: short video, animated, hosted on your repo. That convinces people why they want to work on your code. | 3 | Readme |
| Use of version control tools | 3 | Git utilized |
| Test cases exist | 3 | Evidence |
| Test cases are routinely executed | 3 | Evidence |
| The files CONTRIBUTING.md lists coding standards and lots of tips on how to extend the system without screwing things up | 3 | Contribution information is up to date. Links are included within the file to point to additional resources or areas on the git page that should be used when contributing. |
| Issues are discussed before they are closed | 3 | Evidence |
| Chat channel: exists | 3 | Chat channel is available and is activly used. Evidence |
| Test cases: a large proportion of the issues related to handling failing cases. | 3 | GH Action |
| Evidence that the whole team is using the same tools: everyone can get to all tools and files | 3 | All members contributing across branches. Development/testing has occured on multiple machines. |
| Evidence that the whole team is using the same tools (e.g. config files in the repo, updated by lots of different people) | 3 | Evidence |
| Evidence that the whole team is using the same tools (e.g. tutor can ask anyone to share screen, they demonstrate the system running on their computer) | 3 | The entire team has access to the repository and is able to run the demo. We ensured this by having different members screen share and run the program during troubleshooting sessions. |
| Evidence that the members of the team are working across multiple places in the code base | 3 | All team members worked on the front and the backend |
| Short release cycles | 3 | Evident from multiple commits. |
What's inside
30 scoring rows, each with notes, score, and evidence link
Change this for your project
- Replace
dnellur4/wolfcarev2with your own repository name in all evidence URLs - Replace
wolfcarev2in paths like.github/workflows/python-app.ymlwith your repo name - Replace
symptoms_calculator.htmlwith your own documentation page name
Where it goes
Keep it in your repository where the agent or team that needs it will read it.
Worth borrowing
- Links each rubric criterion directly to concrete evidence in the repo (commits, issues, actions)
- Uses a single table to combine scoring, notes, and evidence for quick review
Related Documents
RUBRIC.md
AICA PRD Quality Rubric
Defines a 10-dimension scoring rubric (0-5 each) to evaluate and create PRDs, with a target score of 36+ for agent-driven implementation.
aiagentllm
0
4
lucasblimaRUBRIC.md
Judging Rubric
Defines a 100-point scoring rubric for an AI-for-social-good hackathon, covering five criteria with detailed score bands and tie-breaking rules.
aievalworkflow
0
3
rudra496RUBRIC.md
Liscribe v2 — Rubric of Success
Defines the complete UI specification, success criteria, and non-negotiable principles for a macOS transcription app with three independent workflows.
airagworkflow
0
3
LenniottRUBRIC.md
QualRubric
Defines a qualitative rubric for PhD qualifying exam presentations, focusing on context, organization, and knowledge demonstration rather than numerical scoring.
aieval
0
3
adsarwate