gen-report-detailed

Generate detailed 13-section markdown experiment reports from flame_moonshot training logs and metrics.

7|1|Updated Apr 1, 2026

Search Tags:#metrics #visualization #reproducibility #ml #failure-analysis #training-logs #experiment-report

Ready to run

npx skills add https://github.com/gyy0592/claude-config --skill gen-report-detailed

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gen-report-detailed
Source: https://github.com/gyy0592/claude-config/tree/main/skills/gen-report-detailed
Command: npx skills add https://github.com/gyy0592/claude-config --skill gen-report-detailed

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the creation of a complete, reproducible 13-section experiment report for flame_moonshot training runs by collecting logs, configs, metrics, artifacts, git metadata, and visualizations so researchers can verify results and diagnose failures efficiently.

Core Features & Use Cases

Collects training logs, CSV metrics, model and run configuration files, SLURM or run scripts, artifacts notes, and available PNG visualizations.
Enforces metric provenance by recording file paths, line numbers and step indices and includes formal mathematical definitions for reported metrics, plus a reproducibility checklist and failure analysis.
Use cases include post-run forensic analysis, preparation of reproducibility bundles for publication, and rapid triage of training anomalies.

Quick Start

Use gen-report-detailed to generate a timestamped detailed markdown report for a given experiment by providing the experiment name and root directory when prompted.

Frequently Asked Questions about gen-report-detailed

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate a reproducible experiment report for ML training logs?▼

To generate a reproducible experiment report, provide the experiment name and root directory to collect training logs, CSV metrics, model configs, and git metadata, producing a timestamped markdown file with metric provenance and failure analysis.

What is metric provenance and how does it work in experiment reports?▼

Metric provenance in experiment reports records file paths, line numbers, and step indices alongside formal mathematical definitions, ensuring researchers can trace reported metrics back to their source data for full reproducibility.

How do I perform failure analysis on training anomalies using CSV metrics and logs?▼

Failure analysis on training anomalies uses collected CSV metrics, training logs, and artifact summaries to diagnose irregularities, producing a structured 13-section report with a reproducibility checklist and detailed findings.

Can I include SLURM run scripts and PNG visualizations in my training report?▼

Yes, the report generation process collects SLURM or run scripts and available PNG visualization files alongside model configurations and artifacts, integrating them into the final timestamped markdown report for comprehensive documentation.

What's the best way to prepare a reproducibility bundle for publication from training artifacts?▼

The best way to prepare a reproducibility bundle is to collect training logs, CSV metrics, git metadata, and visualizations, enforcing metric provenance with formulas to produce a complete 13-section experiment report.

What file format is used for the generated experiment report?▼

The generated experiment report is written to a timestamped markdown file, containing 13 sections with collected logs, metrics, artifact summaries, and a reproducibility checklist for post-run forensic analysis.