Metis — Data skill for Claude Code
Run as Metis (project research specialist for codebase analysis and documentation) inline in the main session — pipeline Stage 0.
How to install Metis
Installs to ~/.claude/skills/lizardliang-kratos-metis/SKILL.md
mkdir -p ~/.claude/skills/lizardliang-kratos-metis && curl -fsSL https://raw.githubusercontent.com/LizardLiang/kratos/HEAD/commands/metis.md -o ~/.claude/skills/lizardliang-kratos-metis/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Metis does
name: metis description: Run as Metis (project research specialist for codebase analysis and documentation) inline in the main session — pipeline Stage 0 generated: true allowed-tools: Bash(echo:*), Bash(node:*)
!`echo "KRATOS_ROOT=${CLAUDE_PLUGIN_ROOT}"`
!`node "${CLAUDE_PLUGIN_ROOT}/hooks/launch.cjs" agent load metis --resolve --part body`
!`node "${CLAUDE_PLUGIN_ROOT}/hooks/launch.cjs" agent load metis --resolve --part extras`
You ARE Metis for this turn. Adopt the persona, t
Alternatives in Data
- Claudekit Research Tools — Research and analysis tools 1.9k ★
- Claude Code Memory Setup — Up to 71.5x fewer tokens per session on Claude Code with Obsidian + Graphify 947 ★
- Graph Engineering — Graph engineering for AI agents: the 9-stage knowledge-graph pipeline (translated from SEU's graduate course) 452 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Apollo
Run as Apollo (architecture reviewer for technical soundness) inline in the main session — pipeline Stage 5
Ares
Run as Ares (implementation specialist for writing code) inline in the main session — pipeline Stage 7a
Dd Docs
Open the Data Documentation and Diagrams department (reflect stage) and route the request to one of its 20 ato
Nexscope Ecommerce API
Ecommerce data and creative AI APIs for marketplace research, product imagery, and video generation, with docu
Invest Committee
Institutional-grade six-stage stock research pipeline for retail investors, as a Claude Code Agent Skill. Taiw
Econ Paper Studio
End-to-end empirical economics workflow for AI agents Five-stage pipeline (question/design/execute/verify/writ
Related Agents
Jesse
Walter's lab partner. Research, verification, and assessment without consuming main context. Use when delegati
Context Pack Loader
Loads a mode-scoped context pack instead of the full agent-facing reference docs. Operator invokes via "load c
Linker
Linker for the thruuu article writing pipeline. Receives the humanized draft, the Links element from the brief