Research.Implement — Development skill for Claude Code
Work the study queue, build and evaluate, maintain claims and progress, and check consistency before finishing; manuscript tasks require explicit selection.
How to install Research.Implement
Installs to ~/.claude/skills/jiancui-research-research-kit-research.implement/SKILL.md
mkdir -p ~/.claude/skills/jiancui-research-research-kit-research.implement && curl -fsSL https://raw.githubusercontent.com/jiancui-research/research-kit/HEAD/commands/research.implement.md -o ~/.claude/skills/jiancui-research-research-kit-research.implement/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Research.Implement does
description: "Work the study queue, build and evaluate, maintain claims and progress, and check consistency before finishing; manuscript tasks require explicit selection." argument-hint: "task ID or queue section, explicit paper work, or check (e.g. \"T012\", \"eval\", \"paper intro\", \"draft evaluation\", \"check claims\")"
Preparation
Resolve `` from the enclosing installed plugin directory (the skill adapter supplies it), else `${CLAUDE_PLUGIN_ROOT}`, else the enabled OM
Alternatives in Development
- Book To Skill — Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work 25.9k ★
- Sprint Plan — Plan sprints with capacity, story selection, and risk mapping 7.8k ★
- Claude Certified Architect — This repository contains study materials for the Claude Certified Architect — Foundations certification 4.6k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Create Flow
Interactive scaffolding command that creates custom agents, commands, skills, and hooks following project conv
Lab Agent
Research topic-selection platform. Drives the agent already installed on your machine (pi / Claude Code / Code
Reality Check
Full reality check of PAN Wizard — claims vs code vs deployed installs, market and runtime delta, peer compari
Se Trade
Trade Study (CREATE) — evaluate ≥2 options for a Spike/Concept, record the decision in a CR with relation(deci
Loop Executor
Implement tasks from the plan. Evaluate-Loop Step 3.
P Idea Wizard
Generate, evaluate, and implement top project improvement ideas
Related Agents
Knuth
SPECIALIST — Data Architect. PostgreSQL, RuVector graph adapter (AD-49 default), Neo4j fallback, query optimiz
Monitor Agent
Continuously monitors system health, agent status, queue metrics, and workflow progress. Provides real-time ob
Writer Critic
Manuscript critic that reviews paper drafts for structure, claims-evidence alignment, identification fidelity,