hugo0076

6 mods across 1 repository, 2 stars between them.

catch-up

01

hugo0076/research-skills

Skill Claude CodeCodex

Bring Hugo up to speed on the project. Use whenever he asks to be caught up or oriented ("catch me up", "where are we", "what's the status", "what's been happening"). Pattern-match these naturally; do not wait for an explicit skill name.

2 2mo ago A 58 tokens

experiment-log

02

hugo0076/research-skills

Skill Claude CodeCodex

Conventions for maintaining docs/experimentlog.md and docs/claimsstatus.md. Load whenever recording an experiment, a result, an observation, or one of Hugo's decisions.

2 2mo ago A 38 tokens

request-briefing

03

hugo0076/research-skills

Skill Claude CodeCodex

Run the intake interview at the start of a new research direction. Use when a session begins with little or no context, or when Hugo says he wants to start a new direction. Run this before designing or launching any experiments.

2 2mo ago A 50 tokens

rigor

04

hugo0076/research-skills

Skill Claude CodeCodex

The playbook for not fooling yourself. Load for any hypothesis-design or result-interpretation step, before trusting a result, and whenever deciding whether a claim is settled.

2 2mo ago A 38 tokens

setup-repo

05

hugo0076/research-skills

Skill Claude CodeCodex

Scaffold a new research project repository, then run the intake briefing to populate its direction. Use when Hugo starts a new research direction or asks to set up a project repo. Picks the host (Spartan or local), assembles and fills the orchestrator CLAUDE.md, seeds the docs files, sets up local git and pre-commit…

2 2mo ago A 95 tokens

slurm-dispatch

06

hugo0076/research-skills

Skill Claude CodeCodex

How to run compute on Spartan. Submit batch jobs, hold and use the interactive GPU bench, detect job completion, retrieve results, and keep your own allocation alive. Load for anything that touches compute.

2 2mo ago A 45 tokens