obsidian-experiment-log

Maintains canonical experiment and result notes in an Obsidian research vault.

Updated Oct 7, 2022
One-click install
npx skills add https://github.com/tamagusko/linux-cfg --skill obsidian-experiment-log-tamagusko
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: obsidian-experiment-log
Source: https://github.com/tamagusko/linux-cfg/tree/main/dotfiles/claude/skills/obsidian-experiment-log
Command: npx skills add https://github.com/tamagusko/linux-cfg --skill obsidian-experiment-log-tamagusko

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Research projects generate scattered experiment logs, metric dumps, and findings that get lost or duplicated across notes. This Skill keeps experiment and result documentation organized in canonical Obsidian notes so durable findings stay traceable and reusable. ## Core Features & Use Cases - Canonical Experiment Notes: Creates or updates structured notes in Experiments/ with goal, hypothesis, dataset, metrics, status, and next steps. - Result Promotion: Promotes stable findings into Results/ notes with key numbers, evidence, interpretation, and a keep/iterate/discard decision. - Controlled Linking: Links experiments and results to each other and to hub, plan, or daily notes only when it improves the working surface, avoiding note sprawl. - Use Case: After finishing a training run with a new baseline, ask the assistant to log the experiment, update the linked result note with the key metrics, and reference it from today's daily note. ## Quick Start Log this training run as an experiment note in Obsidian and update the linked result note with the final metrics.

Frequently Asked Questions about obsidian-experiment-log

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

FAQPage Schema
How do I log machine learning experiments in Obsidian?

Create or update a canonical note in the Experiments folder with sections for goal, hypothesis, code entrypoint, dataset, metrics, status, findings, and next step. Prefer updating an existing experiment note over creating a sibling note for the same experiment line.

When should an experiment result get its own note?

A result note should exist only when the outcome is stable enough to reference later. Raw logs and metric dumps stay in Daily notes until interpreted, then durable findings are promoted into the Results folder with evidence and a keep, iterate, or discard decision.

How do I avoid duplicate experiment notes in Obsidian?

Update the existing canonical experiment note rather than creating a parallel page for the same experiment line. Link experiments and results directly to each other, and only link to hub, plan, or daily notes when it improves the main working surface.

What sections should an experiment note contain?

Minimum sections are goal or hypothesis, code or config entrypoint, dataset and split, metrics, status (planned, running, done, failed), findings, and next step. Result notes additionally record the linked experiment, key numbers, evidence, interpretation, and decision.

Does this work with paper reading and writing workflows?

Yes, experiment notes act as the bridge between Papers and Results folders. Paper-derived hypotheses and baselines land in experiment notes, stable findings are promoted to Results, and claim-worthy results update the Writing folder.