Skills for Real Engineers. Straight from my .agents directory.
nanoresearch-planning
Agent skill fromOpenRaiser/NanoResearch
Produce an experiment blueprint from a research hypothesis
How to install nanoresearch-planning
npx skills add OpenRaiser/NanoResearch# or add only this skill to Claude Code: copy the folder
skills/nanoresearch-planning/
# into
~/.claude/skills/nanoresearch-planning/ (all projects)
.claude/skills/nanoresearch-planning/ (one project)
1# Planning Skill23## Purpose4Take the selected hypothesis from ideation and produce a detailed experiment blueprint specifying datasets, baselines, evaluation metrics, and ablation groups.56## Tools Required7None. This skill operates entirely through LLM reasoning over the ideation output.89## Input10- `ideation_output`: Path to `papers/ideation_output.json` produced by the ideation skill1112## Process131. Parse the selected hypothesis and supporting literature from the ideation output142. Identify candidate datasets that are publicly available and appropriate for validating the hypothesis153. Select 2-4 baseline methods from the surveyed literature for comparison164. Define primary and secondary evaluation metrics aligned with the hypothesis175. Design ablation groups that isolate each novel component of the proposed approach186. Estimate computational requirements and timeline for each experiment197. Compile everything into a structured experiment blueprint2021## Output22Produces `papers/experiment_blueprint.json` containing:23- Selected hypothesis (carried forward)24- Dataset specifications (name, source, splits, preprocessing steps)25- Baseline methods with references26- Evaluation metrics and success criteria27- Ablation study design (groups, variables, expected outcomes)28- Resource estimates and experiment schedule
Excerpt shown under the repository's MIT license.
More skills in NanoResearch
Similar skill repositories
An agentic skills framework & software development methodology that works.
Public repository for Agent Skills
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Production-grade engineering skills for AI coding agents.
