Implement data science analysis tasks with output-first verification and sub-agent delegation.
npx skills add https://github.com/edwinhu/workflows --skill ds-implement
Please help me install this Agent Skill. Skill: ds-implement Source: https://github.com/edwinhu/workflows/tree/main/lib/skills/ds-implement Command: npx skills add https://github.com/edwinhu/workflows --skill ds-implement
This Skill ensures that data science analysis tasks are implemented rigorously, with every step producing verifiable output, preventing silent errors and ensuring reproducibility.
Implement the analysis tasks defined in PLAN.md using the ds-implement skill.
High-intent search queries and answers about installing and using this skill.
An output-first verification protocol enforces data science implementation rigor by mandating visible, verifiable output for every code execution. This prevents silent errors by actively blocking rationalization for skipped checks before proceeding to the next step.
An output-first verification protocol is a mechanism for data science tasks that requires every code execution to produce visible output. It prevents assumptions about correctness by actively guarding against rationalization for skipped checks during analysis implementation.
Yes, you can delegate data analysis tasks to sub-agents during ETL strategy implementation. A delegation protocol separates orchestration from execution, ensuring sub-agents handle analysis while the main workflow enforces output-first verification.
Yes, the data science implementation workflow supports SAS language routing. It enforces ETL strategy based on PLAN.md while routing SAS code and mandating visible output verification for every executed analysis step.
To stop silent errors and rationalization during data transformations, enforce an output-first verification protocol. This actively guards against skipped checks by requiring visible, verifiable output before proceeding, ensuring reliable and reproducible results.
An output-first verification protocol for data analysis requires visible output for every code execution, which limits skipping rapid prototyping steps. It mandates strict delegation to sub-agents and PLAN.md adherence, potentially slowing iterative exploratory analysis.
futureteaCoordinate multi-cluster Rancher management by routing tasks to sub-agents.
tonone-aiAutomate end-to-end ML pipelines from data ingestion to serving endpoints.
erichowensAutomate ETL/ELT pipeline design, validation, and operation across medallion architectures.
Ming-Kai-LCExecute Jupyter notebooks and verify outputs for errors and quality metrics.
