Result reporter
Skill maximussthegreat/ml-researcher-os/skills/result-reporter
Agent skills and workflows for reproducible ML research.
npx -y skills add maximussthegreat/ml-researcher-os --skill result-reporterAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when summarizing experiment logs, benchmark outputs, ablations, failed runs, or model results into an honest research report.
SKILL.md
1.3 KB, as published. Nobody here has run it
Result Reporter
Use this skill after experiments produce logs, metrics, or artifacts.
Goal
Turn experiment outputs into an honest report that separates measured evidence from interpretation.
Required sections
- Summary
- Setup
- Results table
- Observations
- Interpretation
- Limitations
- Negative results
- Next experiments
Evidence tags
Every important statement must be tagged:
[observed]direct from logs or artifacts[computed]derived from available data[cited]from a source[inferred]plausible but not directly measured[speculative]hypothesis for follow-up
Reporting rules
- Do not hide failed runs.
- Do not round away meaningful differences.
- Do not claim state of the art unless the benchmark setup exactly matches.
- Do not compare against a baseline that was not run or cited.
- Include uncertainty when seeds or confidence intervals are missing.
Results table format
| Run | Seed | Dataset split | Metric | Value | Notes |
|---|
Final line
End with a clear answer:
Claim supported: yes/no/partial
Reason: ...