run-report-writer

What problem does it solve?

This Skill eliminates the manual, error-prone process of assembling execution evidence into a consistent, auditable run report after a replication run. It guides the user to capture execution metadata, input validation results, generated outputs, QA gates, encountered issues, and a complete artifact inventory so that runs are reproducible and traceable.

Core Features & Use Cases

• Manual template population: fill the CLAUDE_HANDOFF/RUN_REPORT_TEMPLATE.md using real logs, manifests, and artifacts from a completed run.
• Operational checklist: enumerate execution metadata, input validation summaries, output counts, QA results, and issues encountered.
• Archival practice: save the completed run_report.md in a dated run folder such as 03_outputs/reports/run_YYYYMMDD_HHMM to preserve reproducibility and audit trails.
• Use case: after completing an analysis run, gather the run manifest JSON, execution logs, generated tables and figures, timing/environment details, and produce a consolidated run_report.md for collaborators and publication packages.

Quick Start

Use the run-report-writer to populate CLAUDE_HANDOFF/RUN_REPORT_TEMPLATE.md with evidence from the completed run and save the completed run_report.md into the run folder.