aipoch/open-science-cases

Dernière activité source enregistrée: 16 septembre 2026 à 09:06

Catalogue SkillsMP mis à jour: 9 novembre 2026 à 14:22

skills collectés: 4

Étoiles GitHub: 0
Forks GitHub: 1

Installer avec un assistant IA

Copiez ce prompt dans l’assistant IA que vous utilisez.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/aipoch/open-science-cases.

README du dépôt

Fichier source du README : README.md

Lire le README source

Afficher le README complet

# OpenScience research conversations **Framework 2.2.0** prepares realistic research conversations and an external evidence/assessment workflow. Begin with a scientific question; develop a useful first investigation through native Plan first, execution and runtime auto-review; decide on extensions from the findings. Status: **preparation release; no new live scientific validation claimed**. This repository contains skills, templates, prepared cases and historical preparation only. It contains no scientific run results or application patches. ## Start here - [Framework and roles](docs/framework.md) - [NHANES: smoking and depressive symptoms](cases/nhanes-2015-2016-smoking-phq9/START_Run_Case_with_Codex.md) - [GBM: clinical-trial landscape](cases/gbm-clinical-trial-landscape/START_Run_Case_with_Codex.md) - [Case2: longitudinal COVID-19 proteomics](cases/covid-haemodialysis-proteomics/START_Run_Case_with_Codex.md) - [Case3: TCGA kidney-cancer survival](cases/tcga-kirc-stage-survival/START_Run_Case_with_Codex.md) - [Case2/Case3 migration and local delivery](docs/case2-case3-migration.md) - [Build case preparations](.agents/skills/building-openscience-cases/SKILL.md) - [Operate and assess an authorized case](.agents/skills/openscience-human-review/SKILL.md) - [Migration and installation](docs/migration.md) - [Immutable preparation archive](archive/pre-conversation-20260916/README.md) - [Validation and limitations](docs/validation.md) Before each launch, assign who owns Plan approval, scope questions, extensions, recovery and acceptance. Record authority limits, budget and restart allowance. Preparing or installing these files does not launch a run. Send only a case's `01_OpenScience_Prompt.md` and legitimate scientific context to OpenScience. Keep operating instructions and independent reference work outside the scientific workspace. Configure the current runtime/model through the installed application; this framework carries no machine-specific defaults. ## Maintain and validate Python 3.10+ is sufficient; the repository scripts use the standard library. From the repository root: ```sh python scripts/build_cases.py --check python scripts/validate_preparation.py python -m unittest discover -s tests -v ``` To regenerate after an intentional spec/template/reference edit, run `python scripts/build_cases.py` before checking. Case outputs include source hashes and protocol identities. Frozen archive bytes are never regenerated. Historical release tags: `archive-pre-conversation-20260916` for the preparation baseline; `framework-v2.0.0` for the original conversation release. Report research usefulness, scientific validity, observed product features and Reproduction consistency separately. Consult [CHANGELOG](CHANGELOG.md) for the transition. ## Framework 2.2.0 update The [collection upgrade](docs/collection-migration-2.2.0.md) adds explicit evidence folders, blank records and a portable validator to all four cases. Version 2.2.0 is identified by the framework-v2.2.0 tag; earlier release tags remain historical identities. Actual run evidence and delivery ZIPs remain outside this repository. ## License This project is licensed under the [MIT License](LICENSE).

Chargement du README du dépôt…

Skills dans ce dépôt

Analystes en études de marché et spécialistes en marketing Scientifiques des données

2 catégories métier · 50% classifié

skill

métier

description

mis à jour

building-openscience-cases

Analystes en études de marché et spécialistes en marketing

Prepare OpenScience research cases that begin with a scientific question and develop through native planning, execution and review. Use to create or revise case preparation, not to launch a scientific run.

Langue du texte source : anglais

2026-09-16

openscience-human-review

Scientifiques des données

Operate an authorized OpenScience research conversation through native Plan first, Notebook execution, runtime auto-review and Reproduction, retaining evidence and independently assessing science. Use for case supervision, not for writing the scientific…

Langue du texte source : anglais

2026-09-16

building-openscience-cases

non classé

Prepare or update an OpenScience research case with native Plan first, runtime auto-review, Notebook execution evidence, Reproduction boundaries, and external evaluation under the preserved v0.2 checklist.

Langue du texte source : chinois

2026-09-16

openscience-human-review

non classé

Operate OpenScience cases through native Plan first and runtime auto-review, capture versioned evidence, assess plans and scientific outputs externally, and verify native Reproduction. Use for case supervision and collection, not for authoring the scientific…

Langue du texte source : anglais

2026-09-16

Affichage de 4 skills collectés sur 4.