# Experiment Manager

> Manage lightweight experiment runs by standardizing configs, run logs, metrics, and comparison summaries. Use for reproducible experiment tracking.

- Skill: `lanma-ctrl/experiment-manager` (Agent Skill)
- Install (CLI): `npx skillmds@latest add lanma-ctrl/experiment-manager`
- Raw SKILL.md: https://api.skillmd.com/api/skills/lanma-ctrl/experiment-manager/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: Lanma-ctrl (https://skillmd.com/u/lanma-ctrl)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/lanma-ctrl/experiment-manager

---


# Experiment Manager Skill

Use this skill when running repeated experiments or comparing variants.

## Minimum experiment record
For each run, save:
- run_id
- date/time
- config
- metric(s)
- notes

## File convention
- configs go in `experiments/configs/`
- metrics go in `experiments/results/metrics.csv`
- summaries go in `experiments/reports/`

## Decision rule
Prefer simple CSV/JSON tracking first. Only introduce MLflow/W&B/etc. when the project needs it.