Skill

init

From luc

First-time setup for Luc on a new project. Audits tool availability (muninndb, rtk, context-mode, codebase-memory), installs missing dependencies, configures RTK hooks, injects Luc instructions into CLAUDE.md, and creates the .luc/ project directory. Run once per project — for diagnostics and repair, use /luc:doctor instead.

npx claudepluginhub aperrix/luc

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Set up the Luc tool ecosystem for this project. This skill runs two scripts and displays their results — keep output minimal to save tokens.

Supporting Assets

__init__.pyscripts/__init__.pyscripts/init_audit.pyscripts/init_configure.pyscripts/init_setup.py

SKILL.md

Similar Skills

cache-components

139.3k

Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.

cache-components

mcp-builder

124.2k

Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).

9 files

anthropics-skills-13

Stats

Parent Repo Stars0

Parent Repo Forks0

Last CommitMar 27, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

init

From luc

Luc Init — Audit, Install, and Configure

Set up the Luc tool ecosystem for this project. This skill runs two scripts and displays their results — keep output minimal to save tokens.

Step 1 — Audit

python3 "${CLAUDE_SKILL_DIR}/scripts/init_audit.py"

Parse the output and display a summary table:

Tool

Required

Status

Health

Version

muninndb-lite

yes

ok / missing

pass / fail

>=0.1.0

...

MCP servers are declared in the plugin .mcp.json — they connect automatically when binaries are in PATH.

Report hook status and conflicts below the table. If any conflict has "blocking": true, stop here.

Step 2 — Configure

python3 "${CLAUDE_SKILL_DIR}/scripts/init_configure.py" --project-dir "$CLAUDE_PROJECT_DIR"

This command handles installation and configuration in one shot. Parse the JSON output and display the summary table:

Step

Result

Tools

N/N installed

RTK hook

registered

CLAUDE.md

injected / unchanged

.luc/

created / exists

Vault

name

If summary.required_missing is not empty, warn the user: Required tools still missing: {list}. Run /luc:doctor to diagnose.

If tools were installed during step 2, end with: Restart Claude Code to activate MCP servers.

Rules

Always run the audit first — the configure step uses its results to decide what to install.

Automate everything — the user should not need to do anything manually except restart Claude Code for MCP activation.

The skill is idempotent — running /luc:init twice produces the same result because both scripts check existing state before acting.

Be silent during execution — no narration between steps. Output exactly two things: the audit table and the summary table. Extra commentary wastes context tokens.

Luc Init — Audit, Install, and Configure

Set up the Luc tool ecosystem for this project. This skill runs two scripts and displays their results — keep output minimal to save tokens.

Step 1 — Audit

python3 "${CLAUDE_SKILL_DIR}/scripts/init_audit.py"

Parse the output and display a summary table:

Tool	Required	Status	Health	Version
muninndb-lite	yes	ok / missing	pass / fail	>=0.1.0
...	...	...	...	...

MCP servers are declared in the plugin .mcp.json — they connect automatically when binaries are in PATH.

Report hook status and conflicts below the table. If any conflict has "blocking": true, stop here.

Step 2 — Configure

python3 "${CLAUDE_SKILL_DIR}/scripts/init_configure.py" --project-dir "$CLAUDE_PROJECT_DIR"

This command handles installation and configuration in one shot. Parse the JSON output and display the summary table:

Step	Result
Tools	N/N installed
RTK hook	registered
CLAUDE.md	injected / unchanged
.luc/	created / exists
Vault	name

If summary.required_missing is not empty, warn the user: Required tools still missing: {list}. Run /luc:doctor to diagnose.

If tools were installed during step 2, end with: Restart Claude Code to activate MCP servers.

Rules

Always run the audit first — the configure step uses its results to decide what to install.
Automate everything — the user should not need to do anything manually except restart Claude Code for MCP activation.
The skill is idempotent — running /luc:init twice produces the same result because both scripts check existing state before acting.
Be silent during execution — no narration between steps. Output exactly two things: the audit table and the summary table. Extra commentary wastes context tokens.