run-analysis
Skill0 starsChanged 47 days agoMITView on GitHub
Execute the approved analysis plan as reproducible Python scripts with assumption checks, effect sizes, and publication-quality figures. Use in the analysis phase after GATE 1 approval.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- embryossa/embryo-lab-datatheir own version, first seen Nov 15, 2026
- bohewu/agents_pipelinetheir own version, first seen Jul 11, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- bohewu/agents_pipeline★ 1same name
- embryossa/embryo-lab-data★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- analysis-planSkill
- data-scoutSkill
- literature-scanSkill
- stats-reviewSkill
- write-paperSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Never overwrite raw data. Assert expected dtypes/ranges after cleaning (e.g., age 18–100).
- Distinctive: line 24
Run all pre-specified diagnostics.
- Distinctive: line 38
If results contradict hypotheses, report them exactly as found.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Run Analysis
Execute analysis/plan.md exactly. All code in numbered scripts under analysis/, orchestrated by run_all.py. A fresh run must reproduce every reported number.
Script conventions
00_download.py→data/raw/;01_clean.py→data/processed/analysis_df.parquet;02_descriptives.py;03_main_models.py;04_robustness.py;05_figures.py.- Top of every script:
SEED = 42, set for numpy/random. Print key outputs; also write every estimate the paper will cite (estimate, CI, SE, N, p) toresults/estimates.json. requirements.txtwith pinned versions of used packages.
Cleaning (01)
- Log every step and row-count change: initial N → after each exclusion → final N. Write this flow into
results/results.md(needed for the paper's sample section). - Never overwrite raw data. Assert expected dtypes/ranges after cleaning (e.g., age 18–100); fail loudly on violations.
Descriptives (02)
- Table: N, mean/SD or %, min/max, missingness per analysis variable. Correlation matrix of main variables. Scale reliability if applicable.
Main models (03)
- Exactly as pre-specified. Report per hypothesis: estimate, 95% CI, SE, test statistic, p, effect size, N.
- Run all pre-specified diagnostics. If one fails, run the pre-specified fallback and report BOTH.
- Apply the pre-specified multiplicity correction; report raw and adjusted p.
Robustness (04)
- Each planned check; one summary table: does sign/magnitude/significance of the key estimate hold?
Figures (05)
- matplotlib/seaborn, 300 dpi PNG to
results/figures/. Axis labels with units, N in caption. Key associations: scatter with fit line + CI band (jitter/alpha for large N). Colorblind-safe palette, no chartjunk.
results.md
Per hypothesis: plain-language finding; a statistics sentence ready for the paper (e.g., "r = .21, 95% CI [.15, .27], p < .001, N = 1,204"); figure/table reference; diagnostic outcome; robustness summary. Then a clearly-headed Exploratory section for anything not pre-specified.
Rules
- Numbers in results.md must be programmatically produced (read from estimates.json where possible), never typed from memory.
- If results contradict hypotheses, report them exactly as found. Null results are results.
- Finish by running
python analysis/run_all.pyend-to-end; it must complete without errors.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

